Hydrologic-Process-Based Soil Texture Classifications for Improved Visualization of Landscape Function.

PubMed

Groenendyk, Derek G; Ferré, Ty P A; Thorp, Kelly R; Rice, Amy K

2015-01-01

Soils lie at the interface between the atmosphere and the subsurface and are a key component that control ecosystem services, food production, and many other processes at the Earth's surface. There is a long-established convention for identifying and mapping soils by texture. These readily available, georeferenced soil maps and databases are used widely in environmental sciences. Here, we show that these traditional soil classifications can be inappropriate, contributing to bias and uncertainty in applications from slope stability to water resource management. We suggest a new approach to soil classification, with a detailed example from the science of hydrology. Hydrologic simulations based on common meteorological conditions were performed using HYDRUS-1D, spanning textures identified by the United States Department of Agriculture soil texture triangle. We consider these common conditions to be: drainage from saturation, infiltration onto a drained soil, and combined infiltration and drainage events. Using a k-means clustering algorithm, we created soil classifications based on the modeled hydrologic responses of these soils. The hydrologic-process-based classifications were compared to those based on soil texture and a single hydraulic property, Ks. Differences in classifications based on hydrologic response versus soil texture demonstrate that traditional soil texture classification is a poor predictor of hydrologic response. We then developed a QGIS plugin to construct soil maps combining a classification with georeferenced soil data from the Natural Resource Conservation Service. The spatial patterns of hydrologic response were more immediately informative, much simpler, and less ambiguous, for use in applications ranging from trafficability to irrigation management to flood control. The ease with which hydrologic-process-based classifications can be made, along with the improved quantitative predictions of soil responses and visualization of landscape function, suggest that hydrologic-process-based classifications should be incorporated into environmental process models and can be used to define application-specific maps of hydrologic function.

Development of soil taxation and soil classification as furthered by the Austrian Soil Science Society

NASA Astrophysics Data System (ADS)

Baumgarten, Andreas

2013-04-01

Soil taxation and soil classification are important drivers of soil science in Austria. However, the tasks are quite different: whereas soil taxation aims at the evaluation of the productivity potential of the soil, soil classification focusses on the natural development and - especially nowadays - on functionality of the soil. Since the foundation of the Austrian Soil Science Society (ASSS), representatives both directions of the description of the soil have been involved in the common actions of the society. In the first years it was a main target to improve and standardize field descriptions of the soil. Although both systems differ in the general layout, the experts should comply with identical approaches. According to this work, a lot of effort has been put into the standardization of the soil classification system, thus ensuring a common basis. The development, state of the art and further development of both classification and taxation systems initiated and carried out by the ASSS will be shown.

Automatic Detection of Regions in Spinach Canopies Responding to Soil Moisture Deficit Using Combined Visible and Thermal Imagery

PubMed Central

Raza, Shan-e-Ahmed; Smith, Hazel K.; Clarkson, Graham J. J.; Taylor, Gail; Thompson, Andrew J.; Clarkson, John; Rajpoot, Nasir M.

2014-01-01

Thermal imaging has been used in the past for remote detection of regions of canopy showing symptoms of stress, including water deficit stress. Stress indices derived from thermal images have been used as an indicator of canopy water status, but these depend on the choice of reference surfaces and environmental conditions and can be confounded by variations in complex canopy structure. Therefore, in this work, instead of using stress indices, information from thermal and visible light imagery was combined along with machine learning techniques to identify regions of canopy showing a response to soil water deficit. Thermal and visible light images of a spinach canopy with different levels of soil moisture were captured. Statistical measurements from these images were extracted and used to classify between canopies growing in well-watered soil or under soil moisture deficit using Support Vector Machines (SVM) and Gaussian Processes Classifier (GPC) and a combination of both the classifiers. The classification results show a high correlation with soil moisture. We demonstrate that regions of a spinach crop responding to soil water deficit can be identified by using machine learning techniques with a high accuracy of 97%. This method could, in principle, be applied to any crop at a range of scales. PMID:24892284

Development of an Engineering Soil Database

DTIC Science & Technology

2017-12-27

systems such as agricultural and geological soil classifications and soil parameters. Tier 3 Data were converted into equivalent USCS classification...14 2.7 U.S. Department of Agriculture (USDA) textural soil classification ............................ 16 2.7.1 Properties of USDA textural...Defense ERDC U.S. Army Engineer Research and Development Center ESDB European Soil Database FAO Food and Agriculture Organization (of the United

Hyperspectral Imaging Analysis for the Classification of Soil Types and the Determination of Soil Total Nitrogen

PubMed Central

Jia, Shengyao; Li, Hongyang; Wang, Yanjie; Tong, Renyuan; Li, Qing

2017-01-01

Soil is an important environment for crop growth. Quick and accurately access to soil nutrient content information is a prerequisite for scientific fertilization. In this work, hyperspectral imaging (HSI) technology was applied for the classification of soil types and the measurement of soil total nitrogen (TN) content. A total of 183 soil samples collected from Shangyu City (People’s Republic of China), were scanned by a near-infrared hyperspectral imaging system with a wavelength range of 874–1734 nm. The soil samples belonged to three major soil types typical of this area, including paddy soil, red soil and seashore saline soil. The successive projections algorithm (SPA) method was utilized to select effective wavelengths from the full spectrum. Pattern texture features (energy, contrast, homogeneity and entropy) were extracted from the gray-scale images at the effective wavelengths. The support vector machines (SVM) and partial least squares regression (PLSR) methods were used to establish classification and prediction models, respectively. The results showed that by using the combined data sets of effective wavelengths and texture features for modelling an optimal correct classification rate of 91.8%. could be achieved. The soil samples were first classified, then the local models were established for soil TN according to soil types, which achieved better prediction results than the general models. The overall results indicated that hyperspectral imaging technology could be used for soil type classification and soil TN determination, and data fusion combining spectral and image texture information showed advantages for the classification of soil types. PMID:28974005

Soil classification based on cone penetration test (CPT) data in Western Central Java

NASA Astrophysics Data System (ADS)

Apriyono, Arwan; Yanto, Santoso, Purwanto Bekti; Sumiyanto

2018-03-01

This study presents a modified friction ratio range for soil classification i.e. gravel, sand, silt & clay and peat, using CPT data in Western Central Java. The CPT data was obtained solely from Soil Mechanic Laboratory of Jenderal Soedirman University that covers more than 300 sites within the study area. About 197 data were produced from data filtering process. IDW method was employed to interpolated friction ratio values in a regular grid point for soil classification map generation. Soil classification map was generated and presented using QGIS software. In addition, soil classification map with respect to modified friction ratio range was validated using 10% of total measurements. The result shows that silt and clay dominate soil type in the study area, which is in agreement with two popular methods namely Begemann and Vos. However, the modified friction ratio range produces 85% similarity with laboratory measurements whereby Begemann and Vos method yields 70% similarity. In addition, modified friction ratio range can effectively distinguish fine and coarse grains, thus useful for soil classification and subsequently for landslide analysis. Therefore, modified friction ratio range proposed in this study can be used to identify soil type for mountainous tropical region.

29 CFR Appendix A to Subpart P of... - Soil Classification

Code of Federal Regulations, 2013 CFR

2013-07-01

... 29 Labor 8 2013-07-01 2013-07-01 false Soil Classification A Appendix A to Subpart P of Part 1926..., App. A Appendix A to Subpart P of Part 1926—Soil Classification (a) Scope and application—(1) Scope. This appendix describes a method of classifying soil and rock deposits based on site and environmental...

29 CFR Appendix A to Subpart P of... - Soil Classification

Code of Federal Regulations, 2012 CFR

2012-07-01

... 29 Labor 8 2012-07-01 2012-07-01 false Soil Classification A Appendix A to Subpart P of Part 1926..., App. A Appendix A to Subpart P of Part 1926—Soil Classification (a) Scope and application—(1) Scope. This appendix describes a method of classifying soil and rock deposits based on site and environmental...

29 CFR Appendix A to Subpart P of... - Soil Classification

Code of Federal Regulations, 2014 CFR

2014-07-01

... 29 Labor 8 2014-07-01 2014-07-01 false Soil Classification A Appendix A to Subpart P of Part 1926..., App. A Appendix A to Subpart P of Part 1926—Soil Classification (a) Scope and application—(1) Scope. This appendix describes a method of classifying soil and rock deposits based on site and environmental...

29 CFR Appendix A to Subpart P of... - Soil Classification

Code of Federal Regulations, 2011 CFR

2011-07-01

... 29 Labor 8 2011-07-01 2011-07-01 false Soil Classification A Appendix A to Subpart P of Part 1926..., App. A Appendix A to Subpart P of Part 1926—Soil Classification (a) Scope and application—(1) Scope. This appendix describes a method of classifying soil and rock deposits based on site and environmental...

SoilGrids250m: Global gridded soil information based on machine learning

PubMed Central

Mendes de Jesus, Jorge; Heuvelink, Gerard B. M.; Ruiperez Gonzalez, Maria; Kilibarda, Milan; Blagotić, Aleksandar; Shangguan, Wei; Wright, Marvin N.; Geng, Xiaoyuan; Bauer-Marschallinger, Bernhard; Guevara, Mario Antonio; Vargas, Rodrigo; MacMillan, Robert A.; Batjes, Niels H.; Leenaars, Johan G. B.; Ribeiro, Eloi; Wheeler, Ichsani; Mantel, Stephan; Kempen, Bas

2017-01-01

This paper describes the technical development and accuracy assessment of the most recent and improved version of the SoilGrids system at 250m resolution (June 2016 update). SoilGrids provides global predictions for standard numeric soil properties (organic carbon, bulk density, Cation Exchange Capacity (CEC), pH, soil texture fractions and coarse fragments) at seven standard depths (0, 5, 15, 30, 60, 100 and 200 cm), in addition to predictions of depth to bedrock and distribution of soil classes based on the World Reference Base (WRB) and USDA classification systems (ca. 280 raster layers in total). Predictions were based on ca. 150,000 soil profiles used for training and a stack of 158 remote sensing-based soil covariates (primarily derived from MODIS land products, SRTM DEM derivatives, climatic images and global landform and lithology maps), which were used to fit an ensemble of machine learning methods—random forest and gradient boosting and/or multinomial logistic regression—as implemented in the R packages ranger, xgboost, nnet and caret. The results of 10–fold cross-validation show that the ensemble models explain between 56% (coarse fragments) and 83% (pH) of variation with an overall average of 61%. Improvements in the relative accuracy considering the amount of variation explained, in comparison to the previous version of SoilGrids at 1 km spatial resolution, range from 60 to 230%. Improvements can be attributed to: (1) the use of machine learning instead of linear regression, (2) to considerable investments in preparing finer resolution covariate layers and (3) to insertion of additional soil profiles. Further development of SoilGrids could include refinement of methods to incorporate input uncertainties and derivation of posterior probability distributions (per pixel), and further automation of spatial modeling so that soil maps can be generated for potentially hundreds of soil variables. Another area of future research is the development of methods for multiscale merging of SoilGrids predictions with local and/or national gridded soil products (e.g. up to 50 m spatial resolution) so that increasingly more accurate, complete and consistent global soil information can be produced. SoilGrids are available under the Open Data Base License. PMID:28207752

SoilGrids250m: Global gridded soil information based on machine learning.

PubMed

Hengl, Tomislav; Mendes de Jesus, Jorge; Heuvelink, Gerard B M; Ruiperez Gonzalez, Maria; Kilibarda, Milan; Blagotić, Aleksandar; Shangguan, Wei; Wright, Marvin N; Geng, Xiaoyuan; Bauer-Marschallinger, Bernhard; Guevara, Mario Antonio; Vargas, Rodrigo; MacMillan, Robert A; Batjes, Niels H; Leenaars, Johan G B; Ribeiro, Eloi; Wheeler, Ichsani; Mantel, Stephan; Kempen, Bas

2017-01-01

This paper describes the technical development and accuracy assessment of the most recent and improved version of the SoilGrids system at 250m resolution (June 2016 update). SoilGrids provides global predictions for standard numeric soil properties (organic carbon, bulk density, Cation Exchange Capacity (CEC), pH, soil texture fractions and coarse fragments) at seven standard depths (0, 5, 15, 30, 60, 100 and 200 cm), in addition to predictions of depth to bedrock and distribution of soil classes based on the World Reference Base (WRB) and USDA classification systems (ca. 280 raster layers in total). Predictions were based on ca. 150,000 soil profiles used for training and a stack of 158 remote sensing-based soil covariates (primarily derived from MODIS land products, SRTM DEM derivatives, climatic images and global landform and lithology maps), which were used to fit an ensemble of machine learning methods-random forest and gradient boosting and/or multinomial logistic regression-as implemented in the R packages ranger, xgboost, nnet and caret. The results of 10-fold cross-validation show that the ensemble models explain between 56% (coarse fragments) and 83% (pH) of variation with an overall average of 61%. Improvements in the relative accuracy considering the amount of variation explained, in comparison to the previous version of SoilGrids at 1 km spatial resolution, range from 60 to 230%. Improvements can be attributed to: (1) the use of machine learning instead of linear regression, (2) to considerable investments in preparing finer resolution covariate layers and (3) to insertion of additional soil profiles. Further development of SoilGrids could include refinement of methods to incorporate input uncertainties and derivation of posterior probability distributions (per pixel), and further automation of spatial modeling so that soil maps can be generated for potentially hundreds of soil variables. Another area of future research is the development of methods for multiscale merging of SoilGrids predictions with local and/or national gridded soil products (e.g. up to 50 m spatial resolution) so that increasingly more accurate, complete and consistent global soil information can be produced. SoilGrids are available under the Open Data Base License.

Reflecting on the structure of soil classification systems: insights from a proposal for integrating subsoil data into soil information systems

NASA Astrophysics Data System (ADS)

Dondeyne, Stefaan; Juilleret, Jérôme; Vancampenhout, Karen; Deckers, Jozef; Hissler, Christophe

2017-04-01

Classification of soils in both World Reference Base for soil resources (WRB) and Soil Taxonomy hinges on the identification of diagnostic horizons and characteristics. However as these features often occur within the first 100 cm, these classification systems convey little information on subsoil characteristics. An integrated knowledge of the soil, soil-to-substratum and deeper substratum continuum is required when dealing with environmental issues such as vegetation ecology, water quality or the Critical Zone in general. Therefore, we recently proposed a classification system of the subsolum complementing current soil classification systems. By reflecting on the structure of the subsoil classification system which is inspired by WRB, we aim at fostering a discussion on some potential future developments of WRB. For classifying the subsolum we define Regolite, Saprolite, Saprock and Bedrock as four Subsolum Reference Groups each corresponding to different weathering stages of the subsoil. Principal qualifiers can be used to categorize intergrades of these Subsoil Reference Groups while morphologic and lithologic characteristics can be presented with supplementary qualifiers. We argue that adopting a low hierarchical structure - akin to WRB and in contrast to a strong hierarchical structure as in Soil Taxonomy - offers the advantage of having an open classification system avoiding the need for a priori knowledge of all possible combinations which may be encountered in the field. Just as in WRB we also propose to use principal and supplementary qualifiers as a second level of classification. However, in contrast to WRB we propose to reserve the principal qualifiers for intergrades and to regroup the supplementary qualifiers into thematic categories (morphologic or lithologic). Structuring the qualifiers in this manner should facilitate the integration and handling of both soil and subsoil classification units into soil information systems and calls for paying attention to these structural issues in future developments of WRB.

Correlation of the Rock Mass Rating (RMR) System with the Unified Soil Classification System (USCS): Introduction of the Weak Rock Mass Rating System (W-RMR)

NASA Astrophysics Data System (ADS)

Warren, Sean N.; Kallu, Raj R.; Barnard, Chase K.

2016-11-01

Underground gold mines in Nevada are exploiting increasingly deeper ore bodies comprised of weak to very weak rock masses. The Rock Mass Rating (RMR) classification system is widely used at underground gold mines in Nevada and is applicable in fair to good-quality rock masses, but is difficult to apply and loses reliability in very weak rock mass to soil-like material. Because very weak rock masses are transition materials that border engineering rock mass and soil classification systems, soil classification may sometimes be easier and more appropriate to provide insight into material behavior and properties. The Unified Soil Classification System (USCS) is the most likely choice for the classification of very weak rock mass to soil-like material because of its accepted use in tunnel engineering projects and its ability to predict soil-like material behavior underground. A correlation between the RMR and USCS systems was developed by comparing underground geotechnical RMR mapping to laboratory testing of bulk samples from the same locations, thereby assigning a numeric RMR value to the USCS classification that can be used in spreadsheet calculations and geostatistical analyses. The geotechnical classification system presented in this paper including a USCS-RMR correlation, RMR rating equations, and the Geo-Pick Strike Index is collectively introduced as the Weak Rock Mass Rating System (W-RMR). It is the authors' hope that this system will aid in the classification of weak rock masses and more usable design tools based on the RMR system. More broadly, the RMR-USCS correlation and the W-RMR system help define the transition between engineering soil and rock mass classification systems and may provide insight for geotechnical design in very weak rock masses.

Integrative image segmentation optimization and machine learning approach for high quality land-use and land-cover mapping using multisource remote sensing data

NASA Astrophysics Data System (ADS)

Gibril, Mohamed Barakat A.; Idrees, Mohammed Oludare; Yao, Kouame; Shafri, Helmi Zulhaidi Mohd

2018-01-01

The growing use of optimization for geographic object-based image analysis and the possibility to derive a wide range of information about the image in textual form makes machine learning (data mining) a versatile tool for information extraction from multiple data sources. This paper presents application of data mining for land-cover classification by fusing SPOT-6, RADARSAT-2, and derived dataset. First, the images and other derived indices (normalized difference vegetation index, normalized difference water index, and soil adjusted vegetation index) were combined and subjected to segmentation process with optimal segmentation parameters obtained using combination of spatial and Taguchi statistical optimization. The image objects, which carry all the attributes of the input datasets, were extracted and related to the target land-cover classes through data mining algorithms (decision tree) for classification. To evaluate the performance, the result was compared with two nonparametric classifiers: support vector machine (SVM) and random forest (RF). Furthermore, the decision tree classification result was evaluated against six unoptimized trials segmented using arbitrary parameter combinations. The result shows that the optimized process produces better land-use land-cover classification with overall classification accuracy of 91.79%, 87.25%, and 88.69% for SVM and RF, respectively, while the results of the six unoptimized classifications yield overall accuracy between 84.44% and 88.08%. Higher accuracy of the optimized data mining classification approach compared to the unoptimized results indicates that the optimization process has significant impact on the classification quality.

Classification of anthropogenic soils by new diagnostic criteria involved in the Slovak Soil Classification System (2014)

NASA Astrophysics Data System (ADS)

Sobocká, Jaroslava; Balkovič, Juraj; Bedrna, Zoltán

2017-04-01

Anthropogenic soils can be found mostly in SUITMA areas. The issue of adequate and correct description and classification of these soils occurs very often and can result in inconsistent even in contradictory opinions. In the new version of the anthropogenic soil classification system in Slovakia some new diagnostics criteria were involved and applied for better understanding the inherent nature of these soils. The group of the former anthropogenic soils was divided following scheme of soil reference groups in the WRB 2014 (Anthrozem and Technozem). According to the new version of the Slovak anthropogenic soils classification (2014) there have been distinguished 2 groups of anthropogenic soils: 1) cultivated soils group including 2 soil types (in Slovak terminology): Kultizem and Hortizem and 2) technogenic soils group having 2 soil types: Antrozem and Technozem. Cultivated soil group represents soils developing or forming "in-situ" with diagnostic horizons characterized by human deeply influenced cultivated processes. Technogenic soil group are soils developing like "ex-situ" soils. The key features recognizing technogenic soil group are human-transported and altered material (HTAM = ex-situ aspect), and artefacts content. Diagnostic horizons (top and subsoil) were described as various material affected by physical-mechanical excavation, transportation and spread, mixing, and containing artefacts (the new diagnostic feature). Kultizems are differentiated by cultivated horizon(s) and Technozems by anthropogenic horizon(s). Cultivated horizons are mostly well-known described horizon in many scientific references. Anthropogenic horizons for Technozem are developed from the human-induced transported and altered material which origin is from the other ecological locality that adjacent area. Materials (or substrates) can consist of various material (natural, technogenic or their mixing) with thickness ≥ 60 cm. Artefacts are the second diagnostic feature which presence authenticates the "artificial origin" of the soil. Natural material contains ≤ 10 % artefacts; natural-technogenic 10-40 % artefacts; and technogenic ≥ 40 %. In the soil survey anthropogenic transported or altered layer is very simply recognizable in soil profile if it is compared with adjacent natural horizons. The classification problem is to define and distinguish not only artefacts in soil profile but recognize the origin of the material. The completed manual for these issues is missing. In the contribution, there graphically individual basic soil types of Antrozems and Technozems with some subtypes will be illustrated. Also the basic schema of classification units in Slovakia will be depicted.

The Soil Series in Soil Classifications of the United States

NASA Astrophysics Data System (ADS)

Indorante, Samuel; Beaudette, Dylan; Brevik, Eric C.

2014-05-01

Organized national soil survey began in the United States in 1899, with soil types as the units being mapped. The soil series concept was introduced into the U.S. soil survey in 1903 as a way to relate soils being mapped in one area to the soils of other areas. The original concept of a soil series was all soil types formed in the same parent materials that were of the same geologic age. However, within about 15 years soil series became the primary units being mapped in U.S. soil survey. Soil types became subdivisions of soil series, with the subdivisions based on changes in texture. As the soil series became the primary mapping unit the concept of what a soil series was also changed. Instead of being based on parent materials and geologic age, the soil series of the 1920s was based on the morphology and composition of the soil profile. Another major change in the concept of soil series occurred when U.S. Soil Taxonomy was released in 1975. Under Soil Taxonomy, the soil series subdivisions were based on the uses the soils might be put to, particularly their agricultural uses (Simonson, 1997). While the concept of the soil series has changed over the years, the term soil series has been the longest-lived term in U.S. soil classification. It has appeared in every official classification system used by the U.S. soil survey (Brevik and Hartemink, 2013). The first classification system was put together by Milton Whitney in 1909 and had soil series at its second lowest level, with soil type at the lowest level. The second classification system used by the U.S. soil survey was developed by C.F. Marbut, H.H. Bennett, J.E. Lapham, and M.H. Lapham in 1913. It had soil series at the second highest level, with soil classes and soil types at more detailed levels. This was followed by another system in 1938 developed by M. Baldwin, C.E. Kellogg, and J. Thorp. In this system soil series were again at the second lowest level with soil types at the lowest level. The soil type concept was dropped and replaced by the soil phase in the 1950s in a modification of the 1938 Baldwin et al. classification (Simonson, 1997). When Soil Taxonomy was released in 1975, soil series became the most detailed (lowest) level of the classification system, and the only term maintained throughout all U.S. classifications to date. While the number of recognized soil series have increased steadily throughout the history of U.S. soil survey, there was a rapid increase in the recognition of new soil series following the introduction of Soil Taxonomy (Brevik and Hartemink, 2013). References Brevik, E.C., and A.E. Hartemink. 2013. Soil maps of the United States of America. Soil Science Society of America Journal 77:1117-1132. doi:10.2136/sssaj2012.0390. Simonson, R.W. 1997. Evolution of soil series and type concepts in the United States. Advances in Geoecology 29:79-108.

Data-driven modeling of hydroclimatic trends and soil moisture: Multi-scale data integration and decision support

NASA Astrophysics Data System (ADS)

Coopersmith, Evan Joseph

The techniques and information employed for decision-making vary with the spatial and temporal scope of the assessment required. In modern agriculture, the farm owner or manager makes decisions on a day-to-day or even hour-to-hour basis for dozens of fields scattered over as much as a fifty-mile radius from some central location. Following precipitation events, land begins to dry. Land-owners and managers often trace serpentine paths of 150+ miles every morning to inspect the conditions of their various parcels. His or her objective lies in appropriate resource usage -- is a given tract of land dry enough to be workable at this moment or would he or she be better served waiting patiently? Longer-term, these owners and managers decide upon which seeds will grow most effectively and which crops will make their operations profitable. At even longer temporal scales, decisions are made regarding which fields must be acquired and sold and what types of equipment will be necessary in future operations. This work develops and validates algorithms for these shorter-term decisions, along with models of national climate patterns and climate changes to enable longer-term operational planning. A test site at the University of Illinois South Farms (Urbana, IL, USA) served as the primary location to validate machine learning algorithms, employing public sources of precipitation and potential evapotranspiration to model the wetting/drying process. In expanding such local decision support tools to locations on a national scale, one must recognize the heterogeneity of hydroclimatic and soil characteristics throughout the United States. Machine learning algorithms modeling the wetting/drying process must address this variability, and yet it is wholly impractical to construct a separate algorithm for every conceivable location. For this reason, a national hydrological classification system is presented, allowing clusters of hydroclimatic similarity to emerge naturally from annual regime curve data and facilitate the development of cluster-specific algorithms. Given the desire to enable intelligent decision-making at any location, this classification system is developed in a manner that will allow for classification anywhere in the U.S., even in an ungauged basin. Daily time series data from 428 catchments in the MOPEX database are analyzed to produce an empirical classification tree, partitioning the United States into regions of hydroclimatic similarity. In constructing a classification tree based upon 55 years of data, it is important to recognize the non-stationary nature of climate data. The shifts in climatic regimes will cause certain locations to shift their ultimate position within the classification tree, requiring decision-makers to alter land usage, farming practices, and equipment needs, and algorithms to adjust accordingly. This work adapts the classification model to address the issue of regime shifts over larger temporal scales and suggests how land-usage and farming protocol may vary from hydroclimatic shifts in decades to come. Finally, the generalizability of the hydroclimatic classification system is tested with a physically-based soil moisture model calibrated at several locations throughout the continental United States. The soil moisture model is calibrated at a given site and then applied with the same parameters at other sites within and outside the same hydroclimatic class. The model's performance deteriorates minimally if the calibration and validation location are within the same hydroclimatic class, but deteriorates significantly if the calibration and validates sites are located in different hydroclimatic classes. These soil moisture estimates at the field scale are then further refined by the introduction of LiDAR elevation data, distinguishing faster-drying peaks and ridges from slower-drying valleys. The inclusion of LiDAR enabled multiple locations within the same field to be predicted accurately despite non-identical topography. This cross-application of parametric calibrations and LiDAR-driven disaggregation facilitates decision-support at locations without proximally-located soil moisture sensors.

Keys to soil taxonomy by soil survey staff (sixth edition)

DOE Office of Scientific and Technical Information (OSTI.GOV)

NONE

1994-12-31

This publication, Keys to Soil Taxonomy, serves two purposes. It provides the taxonomic keys necessary for the classification of soils according to Soil Taxonomy in a form that can be used easily in the field, and it also acquaints users of Soil Taxonomy with recent changes in the classification system. This volume includes all revisions of the keys that have so far been approved, replacing the original keys in Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys (1975), the work on which this abridged version, first published in 1983, is based. This publication incorporatesmore » all amendments approved to date and published in National Soil Taxonomy Handbook (NSTH) Issues 1-17.« less

Soil Classification and Treatment.

ERIC Educational Resources Information Center

Clemson Univ., SC. Vocational Education Media Center.

This instructional unit was designed to enable students, primarily at the secondary level, to (1) classify soils according to current capability classifications of the Soil Conservation Service, (2) select treatments needed for a given soil class according to current recommendations provided by the Soil Conservation Service, and (3) interpret a…

Machine learning for predicting soil classes in three semi-arid landscapes

USGS Publications Warehouse

Brungard, Colby W.; Boettinger, Janis L.; Duniway, Michael C.; Wills, Skye A.; Edwards, Thomas C.

2015-01-01

Mapping the spatial distribution of soil taxonomic classes is important for informing soil use and management decisions. Digital soil mapping (DSM) can quantitatively predict the spatial distribution of soil taxonomic classes. Key components of DSM are the method and the set of environmental covariates used to predict soil classes. Machine learning is a general term for a broad set of statistical modeling techniques. Many different machine learning models have been applied in the literature and there are different approaches for selecting covariates for DSM. However, there is little guidance as to which, if any, machine learning model and covariate set might be optimal for predicting soil classes across different landscapes. Our objective was to compare multiple machine learning models and covariate sets for predicting soil taxonomic classes at three geographically distinct areas in the semi-arid western United States of America (southern New Mexico, southwestern Utah, and northeastern Wyoming). All three areas were the focus of digital soil mapping studies. Sampling sites at each study area were selected using conditioned Latin hypercube sampling (cLHS). We compared models that had been used in other DSM studies, including clustering algorithms, discriminant analysis, multinomial logistic regression, neural networks, tree based methods, and support vector machine classifiers. Tested machine learning models were divided into three groups based on model complexity: simple, moderate, and complex. We also compared environmental covariates derived from digital elevation models and Landsat imagery that were divided into three different sets: 1) covariates selected a priori by soil scientists familiar with each area and used as input into cLHS, 2) the covariates in set 1 plus 113 additional covariates, and 3) covariates selected using recursive feature elimination. Overall, complex models were consistently more accurate than simple or moderately complex models. Random forests (RF) using covariates selected via recursive feature elimination was consistently the most accurate, or was among the most accurate, classifiers between study areas and between covariate sets within each study area. We recommend that for soil taxonomic class prediction, complex models and covariates selected by recursive feature elimination be used. Overall classification accuracy in each study area was largely dependent upon the number of soil taxonomic classes and the frequency distribution of pedon observations between taxonomic classes. Individual subgroup class accuracy was generally dependent upon the number of soil pedon observations in each taxonomic class. The number of soil classes is related to the inherent variability of a given area. The imbalance of soil pedon observations between classes is likely related to cLHS. Imbalanced frequency distributions of soil pedon observations between classes must be addressed to improve model accuracy. Solutions include increasing the number of soil pedon observations in classes with few observations or decreasing the number of classes. Spatial predictions using the most accurate models generally agree with expected soil–landscape relationships. Spatial prediction uncertainty was lowest in areas of relatively low relief for each study area.

A Hybrid Semi-supervised Classification Scheme for Mining Multisource Geospatial Data

DOE Office of Scientific and Technical Information (OSTI.GOV)

Vatsavai, Raju; Bhaduri, Budhendra L

2011-01-01

Supervised learning methods such as Maximum Likelihood (ML) are often used in land cover (thematic) classification of remote sensing imagery. ML classifier relies exclusively on spectral characteristics of thematic classes whose statistical distributions (class conditional probability densities) are often overlapping. The spectral response distributions of thematic classes are dependent on many factors including elevation, soil types, and ecological zones. A second problem with statistical classifiers is the requirement of large number of accurate training samples (10 to 30 |dimensions|), which are often costly and time consuming to acquire over large geographic regions. With the increasing availability of geospatial databases, itmore » is possible to exploit the knowledge derived from these ancillary datasets to improve classification accuracies even when the class distributions are highly overlapping. Likewise newer semi-supervised techniques can be adopted to improve the parameter estimates of statistical model by utilizing a large number of easily available unlabeled training samples. Unfortunately there is no convenient multivariate statistical model that can be employed for mulitsource geospatial databases. In this paper we present a hybrid semi-supervised learning algorithm that effectively exploits freely available unlabeled training samples from multispectral remote sensing images and also incorporates ancillary geospatial databases. We have conducted several experiments on real datasets, and our new hybrid approach shows over 25 to 35% improvement in overall classification accuracy over conventional classification schemes.« less

Soil geomorphic classification, soil taxonomy, and effects on soil richness assessments

Treesearch

Jonathan D. Phillips; Daniel A. Marion

2007-01-01

The study of pedodiversity and soil richness depends on the notion of soils as discrete entities. Soil classifications are often criticized in this regard because they depend in part on arbitrary or subjective criteria. In this study soils were categorized on the basis of the presence or absence of six lithological and morphological characteristics. Richness vs. area...

Soil Genesis and Development, Lesson 5 - Soil Geography and Classification

USDA-ARS?s Scientific Manuscript database

The system of soil classification developed by the United States Department of Agriculture (USDA) is called Soil Taxonomy. Soil Taxonomy consists of a hierarchy of six levels which, from highest to lowest, are: Order, Suborder, Great Group, Subgroup, family, and series. This lesson will focus on bro...

[Extracting black soil border in Heilongjiang province based on spectral angle match method].

PubMed

Zhang, Xin-Le; Zhang, Shu-Wen; Li, Ying; Liu, Huan-Jun

2009-04-01

As soils are generally covered by vegetation most time of a year, the spectral reflectance collected by remote sensing technique is from the mixture of soil and vegetation, so the classification precision based on remote sensing (RS) technique is unsatisfied. Under RS and geographic information systems (GIS) environment and with the help of buffer and overlay analysis methods, land use and soil maps were used to derive regions of interest (ROI) for RS supervised classification, which plus MODIS reflectance products were chosen to extract black soil border, with methods including spectral single match. The results showed that the black soil border in Heilongjiang province can be extracted with soil remote sensing method based on MODIS reflectance products, especially in the north part of black soil zone; the classification precision of spectral angel mapping method is the highest, but the classifying accuracy of other soils can not meet the need, because of vegetation covering and similar spectral characteristics; even for the same soil, black soil, the classifying accuracy has obvious spatial heterogeneity, in the north part of black soil zone in Heilongjiang province it is higher than in the south, which is because of spectral differences; as soil uncovering period in Northeastern China is relatively longer, high temporal resolution make MODIS images get the advantage over soil remote sensing classification; with the help of GIS, extracting ROIs by making the best of auxiliary data can improve the precision of soil classification; with the help of auxiliary information, such as topography and climate, the classification accuracy was enhanced significantly. As there are five main factors determining soil classes, much data of different types, such as DEM, terrain factors, climate (temperature, precipitation, etc.), parent material, vegetation map, and remote sensing images, were introduced to classify soils, so how to choose some of the data and quantify the weights of different data layers needs further study.

Classification, Properties, and Management of Aridisols.

ERIC Educational Resources Information Center

Mack, C. B.; And Others

1990-01-01

Described is a slide set which is designed to illustrate the entire range of soils found in the arid regions of the earth's surface. Information on physical and chemical soil properties, soil classification, and related soil management considerations for agricultural development are included. (CW)

Using Landsat MSS data with soils information to identify wetland habitats

NASA Technical Reports Server (NTRS)

Ernst, C. L.; Hoffer, R. M.

1981-01-01

A previous study showed that certain fresh water wetland vegetation types can be spectrally separated when a maximum likelihood classification procedure is applied to Landsat spectral data. However, wetland and upland types which have similar vegetative life forms (e.g., upland hardwoods and hardwood swamps) are often confused because of spectral similarity. Therefore, the current investigation attempts to differentiate similar wetland and upland types by combining Landsat multispectral scanner (MSS) data with soils information. The Pigeon River area in northern Indiana used in the earlier study was also employed in this investigation. A layered classification algorithm which combined soils and spectral data was used to generate a wetland classification. The results of the spectral/soils wetland classification are compared to the previous classification that had been based on spectral data alone. The results indicate wetland habitat mapping can be improved by combining soils and other ancillary data with Landsat spectral data.

Bolivian satellite technology program on ERTS natural resources

NASA Technical Reports Server (NTRS)

Brockmann, H. C. (Principal Investigator); Bartoluccic C., L.; Hoffer, R. M.; Levandowski, D. W.; Ugarte, I.; Valenzuela, R. R.; Urena E., M.; Oros, R.

1977-01-01

The author has identified the following significant results. Application of digital classification for mapping land use permitted the separation of units at more specific levels in less time. A correct classification of data in the computer has a positive effect on the accuracy of the final products. Land use unit comparison with types of soils as represented by the colors of the coded map showed a class relation. Soil types in relation to land cover and land use demonstrated that vegetation was a positive factor in soils classification. Groupings of image resolution elements (pixels) permit studies of land use at different levels, thereby forming parameters for the classification of soils.

Local soil classification and crop suitability: Implications for the historical land use and soil management in Monti di Trapani (Sicily)

NASA Astrophysics Data System (ADS)

Garcia-Vila, Margarita; Corselli, Rocco; Bonet, María Teresa; Lopapa, Giuseppe; Pillitteri, Valentina; Fereres, Elias

2017-04-01

In the past, the lack of technologies (e.g. synthetic fertilizers) to overcome biophysical limitations has played a central role in land use planning. Thus, landscape management and agronomic practices are reactions to local knowledge and perceptions on natural resources, particularly soil. In the framework of the European research project MEMOLA (FP7), the role of local farmers knowledge and perceptions on soil for the historical land use through the spatial distribution of crops and the various management practices have been assessed in three different areas of Monti di Trapani region (Sicily). The identification of the soil classification systems of farmers and the criteria on which it is based, linked to the evaluation of the farmers' ability to identify and map the different soil types, was a key step. Nevertheless, beyond the comparison of the ethnopedological classification approach versus standard soil classification systems, the study also aims at understanding local soil management and land use decisions. The applied methodology was based on an interdisciplinary approach, combining soil science methods and participatory appraisal tools, particularly: i) semi-structured interviews; ii) soil sampling and analysis; iii) discussion groups; and iv) a workshop with local edafologists and agronomists. A rich local glossary of terms associated with the soil conditions and an own soil classification system have been identified in the region. Also, a detailed soil map, including process of soil degradation and soil capability, has been generated. This traditional soil knowledge has conditioned the management and the spatial distribution of the crops, and therefore the configuration of the landscape, until the 1990s. Acknowledgements This work has been funded by the European Union project MEMOLA (Grant agreement no: 613265).

Uncertainties and Solutions Related to Use of WRB (2007) in the Boreo-nemoral zone, Case of Latvia

NASA Astrophysics Data System (ADS)

Kasparinskis, Raimonds; Nikodemus, Olgerts; Rolavs, Nauris

2014-05-01

Relatively high diversity of soils groups according to the WRB (2007) classification is observed in forest ecosystems in the boreo-nemoral zone in Latvia. This is due to the geological genesis of area and environmental conditions (Kasparinskis, Nikodemus, 2012), as well as historical land use and management (Nikodemus et al., 2013). Due to the relatively young soils, Albic, Spodic and Cambic horizons are relatively weakly expressed in many cases. Relatively well developed Albic horizons occur in sandy forest soils, but unusually well expressed Spodic features are observed. In some cases there is a Cambic horizon, however location of Cambisols in the WRB (2007) soil classification sequence does not provide an opportunity to classify these soils as Cambisols, but they are classified as Arenosols. This sequence does not reflect the logical sheme of soil development, and therefore raises the question about location of Podzols, Arenosols and Cambisols in the sequence of WRB (2007) soil classification. Soils with two parent materials (abrupt textural change) are relatively common in Latvia, where conceptually on the small scale mapping results in classification as the soil group Planosols, but in many cases there is occurrence of Fluvic materials, as parent material in the upper part of the soil profile is formed by Baltic Ice lake sandy sediments - this leads to question about the location of Fluvisols and Planosols in the sequence of the WRB (2007) soil classification. Soil research has found cases, where a relatively well developed Spodic horizon was established as the result of ground water table depth in areas of abrupt textural change. In this case the profile corresponds to the soil group of Podzols, however in some cases - Gleysols not Planosols due to a high ground water table. Therefore there is a need for discussion also about the location of Podzols and Planosols in the sequence of the WRB (2007) soil classification. The above mentioned questions raise problems related to unambiguous determination of soil groups. Soil classification must be very precise by reflecting relationships of soil forming processes. In the development of international soil classification it is advisable to pay more attention on ecological processes. This study was supported by the European Social Fund No. 2013/0020/1DP/1.1.1.2.0/13/APIA/VIAA/066. References: IUSS Working Group, 2007. World Reference Base for Soil Resources 2006, first update 2007. World Soil Resources Reports 103. FAO, Rome. 103-116. Kasparinskis R., Nikodemus O. 2012. Influence of environmental factors on the spatial distribution and diversity of forest soil in Latvia. Estonian Journal of Earth Sciences. 61(1): 48-64. Nikodemus O., Kasparinskis R., Kukuls I. 2013. Influence of Afforestation on Soil Genesis, Morphology and Properties in Glacial Till Deposits. Archives of Agronomy and Soil Science. 59(3): 449-465.

Construction of an Yucatec Maya soil classification and comparison with the WRB framework

PubMed Central

2010-01-01

Background Mayas living in southeast Mexico have used soils for millennia and provide thus a good example for understanding soil-culture relationships and for exploring the ways indigenous people name and classify the soils of their territory. This paper shows an attempt to organize the Maya soil knowledge into a soil classification scheme and compares the latter with the World Reference Base for Soil Resources (WRB). Methods Several participative soil surveys were carried out in the period 2000-2009 with the help of bilingual Maya-Spanish-speaking farmers. A multilingual soil database was built with 315 soil profile descriptions. Results On the basis of the diagnostic soil properties and the soil nomenclature used by Maya farmers, a soil classification scheme with a hierarchic, dichotomous and open structure was constructed, organized in groups and qualifiers in a fashion similar to that of the WRB system. Maya soil properties were used at the same categorical levels as similar diagnostic properties are used in the WRB system. Conclusions The Maya soil classification (MSC) is a natural system based on key properties, such as relief position, rock types, size and quantity of stones, color of topsoil and subsoil, depth, water dynamics, and plant-supporting processes. The MSC addresses the soil properties of surficial and subsurficial horizons, and uses plant communities as qualifier in some cases. The MSC is more accurate than the WRB for classifying Leptosols. PMID:20152047

Construction of an Yucatec Maya soil classification and comparison with the WRB framework.

PubMed

Bautista, Francisco; Zinck, J Alfred

2010-02-13

Mayas living in southeast Mexico have used soils for millennia and provide thus a good example for understanding soil-culture relationships and for exploring the ways indigenous people name and classify the soils of their territory. This paper shows an attempt to organize the Maya soil knowledge into a soil classification scheme and compares the latter with the World Reference Base for Soil Resources (WRB). Several participative soil surveys were carried out in the period 2000-2009 with the help of bilingual Maya-Spanish-speaking farmers. A multilingual soil database was built with 315 soil profile descriptions. On the basis of the diagnostic soil properties and the soil nomenclature used by Maya farmers, a soil classification scheme with a hierarchic, dichotomous and open structure was constructed, organized in groups and qualifiers in a fashion similar to that of the WRB system. Maya soil properties were used at the same categorical levels as similar diagnostic properties are used in the WRB system. The Maya soil classification (MSC) is a natural system based on key properties, such as relief position, rock types, size and quantity of stones, color of topsoil and subsoil, depth, water dynamics, and plant-supporting processes. The MSC addresses the soil properties of surficial and subsurficial horizons, and uses plant communities as qualifier in some cases. The MSC is more accurate than the WRB for classifying Leptosols.

Soil classification and carbon storage in cacao agroforestry farming systems of Bahia, Brazil

USDA-ARS?s Scientific Manuscript database

Information concerning the classification of soils and their properties under cacao agroforestry systems of the Atlantic rain forest biome region in the Southeast of Bahia Brazil is largely unknown. Soil and climatic conditions in this region are favorable for high soil carbon storage. This study is...

From landscape to domain: Soils role in landscape classifications

USDA-ARS?s Scientific Manuscript database

Soil landscape classifications are designed to divide landscapes into units with significance for the provisioning and regulating of ecosystem services and the development of conservation plans for natural resources. More specifically, such classifications serve as the basis for stratifying manageme...

Evaluation of automated global mapping of Reference Soil Groups of WRB2015

NASA Astrophysics Data System (ADS)

Mantel, Stephan; Caspari, Thomas; Kempen, Bas; Schad, Peter; Eberhardt, Einar; Ruiperez Gonzalez, Maria

2017-04-01

SoilGrids is an automated system that provides global predictions for standard numeric soil properties at seven standard depths down to 200 cm, currently at spatial resolutions of 1km and 250m. In addition, the system provides predictions of depth to bedrock and distribution of soil classes based on WRB and USDA Soil Taxonomy (ST). In SoilGrids250m(1), soil classes (WRB, version 2006) consist of the RSG and the first prefix qualifier, whereas in SoilGrids1km(2), the soil class was assessed at RSG level. Automated mapping of World Reference Base (WRB) Reference Soil Groups (RSGs) at a global level has great advantages. Maps can be updated in a short time span with relatively little effort when new data become available. To translate soil names of older versions of FAO/WRB and national classification systems of the source data into names according to WRB 2006, correlation tables are used in SoilGrids. Soil properties and classes are predicted independently from each other. This means that the combinations of soil properties for the same cells or soil property-soil class combinations do not necessarily yield logical combinations when the map layers are studied jointly. The model prediction procedure is robust and probably has a low source of error in the prediction of RSGs. It seems that the quality of the original soil classification in the data and the use of correlation tables are the largest sources of error in mapping the RSG distribution patterns. Predicted patterns of dominant RSGs were evaluated in selected areas and sources of error were identified. Suggestions are made for improvement of WRB2015 RSG distribution predictions in SoilGrids. Keywords: Automated global mapping; World Reference Base for Soil Resources; Data evaluation; Data quality assurance References 1 Hengl T, de Jesus JM, Heuvelink GBM, Ruiperez Gonzalez M, Kilibarda M, et al. (2016) SoilGrids250m: global gridded soil information based on Machine Learning. Earth System Science Data (ESSD), in review. 2 Hengl T, de Jesus JM, MacMillan RA, Batjes NH, Heuvelink GBM, et al. (2014) SoilGrids1km — Global Soil Information Based on Automated Mapping. PLoS ONE 9(8): e105992. doi:10.1371/journal.pone.0105992

Soils [Chapter 4.2

Treesearch

Daniel G. Neary; Johannes W. A. Langeveld

2015-01-01

Soils are crucial for profitable and sustainable biomass feedstock production. They provide nutrients and water, give support for plants, and provide habitat for enormous numbers of biota. There are several systems for soil classification. FAO has provided a generic classification system that was used for a global soil map (Bot et al., 2000). The USDA Natural Resources...

Classification problems of Mount Kenya soils

NASA Astrophysics Data System (ADS)

Mutuma, Evans; Csorba, Ádám; Wawire, Amos; Dobos, Endre; Michéli, Erika

2017-04-01

Soil sampling on the agricultural lands covering 1200 square kilometers in the Eastern part of Mount Kenya was carried out to assess the status of soil organic carbon (SOC) as a soil fertility indicator, and to create an up-to-date soil classification map. The geology of the area consists of volcanic rocks and recent superficial deposits. The volcanic rocks are related to the Pliocene time; mainly: lahars, phonolites, tuffs, basalt and ashes. A total of 28 open profiles and 49 augered profiles with 269 samples were collected. The samples were analyzed for total carbon, organic carbon, particle size distribution, percent bases, cation exchange capacity and pH among other parameters. The objective of the study was to evaluate the variability of SOC in different Reference Soil Groups (RGS) and to compare the determined classification units with the KENSOTER database. Soil classification was performed based on the World Reference Base (WRB) for Soil Resources 2014. Based on the earlier surveys, geological and environmental setting, Nitisols were expected to be the dominant soils of the sampled area. However, this was not the case. The major differences to earlier survey data (KENSOTER database) are the presence of high activity clays (CEC value range 27.6 cmol/kg - 70 cmol/kg), high silt content (range 32.6 % - 52.4 %) and silt/clay ratio (range of 0.6 - 1.4) keeping these soils out of the Nitisols RSG. There was good accordance in the morphological features with the earlier survey but failed the silt/clay ratio criteria for Nitisols. This observation calls attention to set new classification criteria for Nitisols and other soils of warm, humid regions with variable rate of weathering to avoid difficulties in interpretation. To address the classification problem, this paper further discusses the taxonomic relationships between the studied soils. On the contrary most of the diagnostic elements (like the presence Umbric horizon, Vitric and Andic properties) and the some qualifiers (Humic, Dystric, Clayic, Skeletic, Leptic, etc) represent useful information for land use and management in the area.

Should there be a "Wet" Soil Order in Soil Taxonomy?

NASA Astrophysics Data System (ADS)

Rabenhorst, Martin; Wessel, Barret; Stolt, Mark; Lindbo, David

2017-04-01

Early soil classification systems recognized wet soils at the highest categorical level. Among the Intrazonal Soils of the US classification utilized between the 1920s and 1960, were included as Great Soil Groups, the Wiesenboden, Bog, Half-Bog, Ground-Water Podzols and Ground-Water Laterites. In other systems, groups named with such terms as ground water gley and pseudogley were also used. With the advent of Soil Taxonomy and it's precursor (1960, 1975), Histosols (organic soils) were distinguished as one of the initial 10 soil orders, and while many of these organic soils are wet soils, some are not (Folists for example). Thus, for over 50 years, with the exception of Histosols, wet soils (which typically represent the wettest end of subaerial wet soils) have not been collectively recognized within taxa at the highest categorical level (order) in the US soil classification system. Rather, the wettest soils were designated at the second categorical level as wet (Aqu) suborders among the various soil orders, and more recently, subaqueous soils as "Wass" suborders of Entisols and Histosols. Soils with less-wet conditions have been recognized at the subgroup (4th) level. Further, in impoundments and regions of transgressing coastlines, submerged upland soils have been found that still classify in soil orders that do not accommodate subaqueous soils ("Wass" suborders). Notwithstanding, other contemporary soil classification systems do (have continued to) recognize wet soils at the highest level. In the World Reference Base (WRB) for example, wet soils are designated as Gleysols or Stagnosols. As efforts are underway to revisit, simplify, and revise Soil Taxonomy, questions have been raised regarding whether wet soils should again be moved back with a place among taxa at the highest category using a name such as Hydrasols, Aquasols, etc. This paper will explore and consider the questions and arguments for and against such proposals and the difficult question regarding where along the soil wetness continuum would be the best point for recognizing a wet soil order.

Estonian soil classification as a tool for recording information on soil cover and its matching with local site types, plant covers and humus forms classifications

NASA Astrophysics Data System (ADS)

Kõlli, Raimo; Tõnutare, Tõnu; Rannik, Kaire; Krebstein, Kadri

2015-04-01

Estonian soil classification (ESC) has been used successfully during more than half of century in soil survey, teaching of soil science, generalization of soil databases, arrangement of soils sustainable management and others. The Estonian normally developed (postlithogenic) mineral soils (form 72.4% from total area) are characterized by mean of genetic-functional schema, where the pedo-ecological position of soils (ie. location among other soils) is given by means of three scalars: (i) 8 stage lithic-genetic scalar (from rendzina to podzols) separates soils each from other by parent material, lithic properties, calcareousness, character of soil processes and others, (ii) 6 stage moisture and aeration conditions scalar (from aridic or well aerated to permanently wet or reductic conditions), and (iii) 2-3 stage soil development scalar, which characterizes the intensity of soil forming processes (accumulation of humus, podzolization). The organic soils pedo-ecological schema, which links with histic postlithogenic soils, is elaborated for characterizing of peatlands superficial mantle (form 23.7% from whole soil cover). The position each peat soil species among others on this organic (peat) soil matrix schema is determined by mean of 3 scalars: (i) peat thickness, (ii) type of paludification or peat forming peculiarities, and (iii) stage of peat decomposition or peat type. On the matrix of abnormally developed (synlithogenic) soils (all together 3.9%) the soil species are positioned (i) by proceeding in actual time geological processes as erosion, fluvial processes (at vicinity of rivers, lakes or sea) or transforming by anthropogenic and technological processes, and (ii) by 7 stage moisture conditions (from aridic to subaqual) of soils. The most important functions of soil cover are: (i) being a suitable environment for plant productivity; (ii) forming adequate conditions for decomposition, transformation and conversion of falling litter (characterized by humus cover type); (iii) being compartment for deposition of humus, individual organic compounds, plant nutrition elements, air and water, and (iv) forming (bio)chemically variegated active space for soil type specific edaphon. For studying of ESC matching with others ecosystem compartments classifications the comparative analysis of corresponding classification schemas was done. It may be concluded that forest and natural grasslands site types as well the plant associations of forests and grasslands correlate (match) well with ESC and therefore these compartments may be adequately expressed on soil cover matrixes. Special interest merits humus cover (in many countries known as humus form), which is by the issue natural body between plant and soil or plant cover and soil cover. The humus cover, which lied on superficial part of soil cover, has been formed by functional interrelationships of plants and soils, reflects very well the local pedo-ecological conditions (both productivity and decomposition cycles) and, therefore, the humus cover types are good indicators for characterizing of local pedo-ecological conditions. The classification of humus covers (humus forms) should be bound with soil classifications. It is important to develop a pedocentric approach in treating of fabric and functioning of natural and agro-ecosystems. Such, based on soil properties, ecosystem approach to management and protection natural resources is highly recommended at least in temperate climatic regions. The sound matching of soil and plant cover is of decisive importance for sustainable functioning of ecosystem and in attaining a good environmental status of the area.

Towards automatic lithological classification from remote sensing data using support vector machines

NASA Astrophysics Data System (ADS)

Yu, Le; Porwal, Alok; Holden, Eun-Jung; Dentith, Michael

2010-05-01

Remote sensing data can be effectively used as a mean to build geological knowledge for poorly mapped terrains. Spectral remote sensing data from space- and air-borne sensors have been widely used to geological mapping, especially in areas of high outcrop density in arid regions. However, spectral remote sensing information by itself cannot be efficiently used for a comprehensive lithological classification of an area due to (1) diagnostic spectral response of a rock within an image pixel is conditioned by several factors including the atmospheric effects, spectral and spatial resolution of the image, sub-pixel level heterogeneity in chemical and mineralogical composition of the rock, presence of soil and vegetation cover; (2) only surface information and is therefore highly sensitive to the noise due to weathering, soil cover, and vegetation. Consequently, for efficient lithological classification, spectral remote sensing data needs to be supplemented with other remote sensing datasets that provide geomorphological and subsurface geological information, such as digital topographic model (DEM) and aeromagnetic data. Each of the datasets contain significant information about geology that, in conjunction, can potentially be used for automated lithological classification using supervised machine learning algorithms. In this study, support vector machine (SVM), which is a kernel-based supervised learning method, was applied to automated lithological classification of a study area in northwestern India using remote sensing data, namely, ASTER, DEM and aeromagnetic data. Several digital image processing techniques were used to produce derivative datasets that contained enhanced information relevant to lithological discrimination. A series of SVMs (trained using k-folder cross-validation with grid search) were tested using various combinations of input datasets selected from among 50 datasets including the original 14 ASTER bands and 36 derivative datasets (including 14 principal component bands, 14 independent component bands, 3 band ratios, 3 DEM derivatives: slope/curvatureroughness and 2 aeromagnetic derivatives: mean and variance of susceptibility) extracted from the ASTER, DEM and aeromagnetic data, in order to determine the optimal inputs that provide the highest classification accuracy. It was found that a combination of ASTER-derived independent components, principal components and band ratios, DEM-derived slope, curvature and roughness, and aeromagnetic-derived mean and variance of magnetic susceptibility provide the highest classification accuracy of 93.4% on independent test samples. A comparison of the classification results of the SVM with those of maximum likelihood (84.9%) and minimum distance (38.4%) classifiers clearly show that the SVM algorithm returns much higher classification accuracy. Therefore, the SVM method can be used to produce quick and reliable geological maps from scarce geological information, which is still the case with many under-developed frontier regions of the world.

Analysis of SURRGO Data and Obtaining Soil Texture Classifications for Simulating Hydrologic Processes

DTIC Science & Technology

2016-07-01

Note (CHETN) describes a method using the U.S. Department of Agriculture (USDA), Natural Resources Conservation Service (NRCS), Soil Survey Geographic...the general texture classifications. 2. Another source for soil information, such as the Food and Agriculture Organization of the United Nations (FAO...science studies such as agriculture , geology, geomorphology, engineering, biology, history, etc. (Soil Survey Division Staff 1993). The procedure pulls

Quality Evaluation of Land-Cover Classification Using Convolutional Neural Network

NASA Astrophysics Data System (ADS)

Dang, Y.; Zhang, J.; Zhao, Y.; Luo, F.; Ma, W.; Yu, F.

2018-04-01

Land-cover classification is one of the most important products of earth observation, which focuses mainly on profiling the physical characters of the land surface with temporal and distribution attributes and contains the information of both natural and man-made coverage elements, such as vegetation, soil, glaciers, rivers, lakes, marsh wetlands and various man-made structures. In recent years, the amount of high-resolution remote sensing data has increased sharply. Accordingly, the volume of land-cover classification products increases, as well as the need to evaluate such frequently updated products that is a big challenge. Conventionally, the automatic quality evaluation of land-cover classification is made through pixel-based classifying algorithms, which lead to a much trickier task and consequently hard to keep peace with the required updating frequency. In this paper, we propose a novel quality evaluation approach for evaluating the land-cover classification by a scene classification method Convolutional Neural Network (CNN) model. By learning from remote sensing data, those randomly generated kernels that serve as filter matrixes evolved to some operators that has similar functions to man-crafted operators, like Sobel operator or Canny operator, and there are other kernels learned by the CNN model that are much more complex and can't be understood as existing filters. The method using CNN approach as the core algorithm serves quality-evaluation tasks well since it calculates a bunch of outputs which directly represent the image's membership grade to certain classes. An automatic quality evaluation approach for the land-cover DLG-DOM coupling data (DLG for Digital Line Graphic, DOM for Digital Orthophoto Map) will be introduced in this paper. The CNN model as an robustness method for image evaluation, then brought out the idea of an automatic quality evaluation approach for land-cover classification. Based on this experiment, new ideas of quality evaluation of DLG-DOM coupling land-cover classification or other kinds of labelled remote sensing data can be further studied.

A discrimlnant function approach to ecological site classification in northern New England

Treesearch

James M. Fincher; Marie-Louise Smith

1994-01-01

Describes one approach to ecologically based classification of upland forest community types of the White and Green Mountain physiographic regions. The classification approach is based on an intensive statistical analysis of the relationship between the communities and soil-site factors. Discriminant functions useful in distinguishing between types based on soil-site...

Spectral band selection for classification of soil organic matter content

NASA Technical Reports Server (NTRS)

Henderson, Tracey L.; Szilagyi, Andrea; Baumgardner, Marion F.; Chen, Chih-Chien Thomas; Landgrebe, David A.

1989-01-01

This paper describes the spectral-band-selection (SBS) algorithm of Chen and Landgrebe (1987, 1988, and 1989) and uses the algorithm to classify the organic matter content in the earth's surface soil. The effectiveness of the algorithm was evaluated comparing the results of classification of the soil organic matter using SBS bands with those obtained using Landsat MSS bands and TM bands, showing that the algorithm was successful in finding important spectral bands for classification of organic matter content. Using the calculated bands, the probabilities of correct classification for climate-stratified data were found to range from 0.910 to 0.980.

Principles of soil mapping of a megalopolis with St. Petersburg as an example

NASA Astrophysics Data System (ADS)

Aparin, B. F.; Sukhacheva, E. Yu.

2014-07-01

For the first time, a soil map of St. Petersburg has been developed on a scale of 1 : 50000 using MicroStation V8i software. The legend to this map contains more than 60 mapping units. The classification of urban soils and information on the soil cover patterns are principally new elements of this legend. New concepts of the urbanized soil space and urbopedocombinations have been suggested for soil mapping of urban territories. The typification of urbopedocombinations in St. Petersburg has been performed on the basis of data on the geometry and composition of the polygons of soils and nonsoil formations. The ratio between the areas of soils and nonsoil formations and their spatial distribution patterns have been used to distinguish between six types of the urbanized soil space. The principles of classification of the soils of urban territories have been specified, and a separate order of pedo-allochthonous soils has been suggested for inclusion into the Classification and Diagnostic System of Russian Soils (2004). Six types of pedo-allochthonous soils have been distinguished on the basis of data on their humus and organic horizons and the character of the underlying mineral substrate.

Fatty acid methyl ester analysis to identify sources of soil in surface water.

PubMed

Banowetz, Gary M; Whittaker, Gerald W; Dierksen, Karen P; Azevedo, Mark D; Kennedy, Ann C; Griffith, Stephen M; Steiner, Jeffrey J

2006-01-01

Efforts to improve land-use practices to prevent contamination of surface waters with soil are limited by an inability to identify the primary sources of soil present in these waters. We evaluated the utility of fatty acid methyl ester (FAME) profiles of dry reference soils for multivariate statistical classification of soils collected from surface waters adjacent to agricultural production fields and a wooded riparian zone. Trials that compared approaches to concentrate soil from surface water showed that aluminum sulfate precipitation provided comparable yields to that obtained by vacuum filtration and was more suitable for handling large numbers of samples. Fatty acid methyl ester profiles were developed from reference soils collected from contrasting land uses in different seasons to determine whether specific fatty acids would consistently serve as variables in multivariate statistical analyses to permit reliable classification of soils. We used a Bayesian method and an independent iterative process to select appropriate fatty acids and found that variable selection was strongly impacted by the season during which soil was collected. The apparent seasonal variation in the occurrence of marker fatty acids in FAME profiles from reference soils prevented preparation of a standardized set of variables. Nevertheless, accurate classification of soil in surface water was achieved utilizing fatty acid variables identified in seasonally matched reference soils. Correlation analysis of entire chromatograms and subsequent discriminant analyses utilizing a restricted number of fatty acid variables showed that FAME profiles of soils exposed to the aquatic environment still had utility for classification at least 1 wk after submersion.

Exploiting machine learning algorithms for tree species classification in a semiarid woodland using RapidEye image

NASA Astrophysics Data System (ADS)

Adelabu, Samuel; Mutanga, Onisimo; Adam, Elhadi; Cho, Moses Azong

2013-01-01

Classification of different tree species in semiarid areas can be challenging as a result of the change in leaf structure and orientation due to soil moisture constraints. Tree species mapping is, however, a key parameter for forest management in semiarid environments. In this study, we examined the suitability of 5-band RapidEye satellite data for the classification of five tree species in mopane woodland of Botswana using machine leaning algorithms with limited training samples.We performed classification using random forest (RF) and support vector machines (SVM) based on EnMap box. The overall accuracies for classifying the five tree species was 88.75 and 85% for both SVM and RF, respectively. We also demonstrated that the new red-edge band in the RapidEye sensor has the potential for classifying tree species in semiarid environments when integrated with other standard bands. Similarly, we observed that where there are limited training samples, SVM is preferred over RF. Finally, we demonstrated that the two accuracy measures of quantity and allocation disagreement are simpler and more helpful for the vast majority of remote sensing classification process than the kappa coefficient. Overall, high species classification can be achieved using strategically located RapidEye bands integrated with advanced processing algorithms.

Working memory supports inference learning just like classification learning.

PubMed

Craig, Stewart; Lewandowsky, Stephan

2013-08-01

Recent research has found a positive relationship between people's working memory capacity (WMC) and their speed of category learning. To date, only classification-learning tasks have been considered, in which people learn to assign category labels to objects. It is unknown whether learning to make inferences about category features might also be related to WMC. We report data from a study in which 119 participants undertook classification learning and inference learning, and completed a series of WMC tasks. Working memory capacity was positively related to people's classification and inference learning performance.

Use of physically-based models and Soil Taxonomy to identify soil moisture classes: Problems and proposals

NASA Astrophysics Data System (ADS)

Bonfante, A.; Basile, A.; de Mascellis, R.; Manna, P.; Terribile, F.

2009-04-01

Soil classification according to Soil Taxonomy include, as fundamental feature, the estimation of soil moisture regime. The term soil moisture regime refers to the "presence or absence either of ground water or of water held at a tension of less than 1500 kPa in the soil or in specific horizons during periods of the year". In the classification procedure, defining of the soil moisture control section is the primary step in order to obtain the soil moisture regimes classification. Currently, the estimation of soil moisture regimes is carried out through simple calculation schemes, such as Newhall and Billaux models, and only in few cases some authors suggest the use of different more complex models (i.e., EPIC) In fact, in the Soil Taxonomy, the definition of the soil moisture control section is based on the wetting front position in two different conditions: the upper boundary is the depth to which a dry soil will be moistened by 2.5 cm of water within 24 hours and the lower boundary is the depth to which a dry soil will be moistened by 7.5 cm of water within 48 hours. Newhall, Billaux and EPIC models don't use physical laws to describe soil water flows, but they use a simple bucket-like scheme where the soil is divided into several compartments and water moves, instantly, only downward when the field capacity is achieved. On the other side, a large number of one-dimensional hydrological simulation models (SWAP, Cropsyst, Hydrus, MACRO, etc..) are available, tested and successfully used. The flow is simulated according to pressure head gradients through the numerical solution of the Richard's equation. These simulation models can be fruitful used to improve the study of soil moisture regimes. The aims of this work are: (i) analysis of the soil moisture control section concept by a physically based model (SWAP); (ii) comparison of the classification obtained in five different Italian pedoclimatic conditions (Mantova and Lodi in northern Italy; Salerno, Benevento and Caserta in southern Italy) applying the classical models (Newhall e Billaux) and the physically-based models (CropSyst e SWAP), The results have shown that the Soil Taxonomy scheme for the definition of the soil moisture regime is unrealistic for the considered Mediterranean soil hydrological conditions. In fact, the same classifications arise irrespective of the soil type. In this respect some suggestions on how modified the section control boundaries were formulated. Keywords: Soil moisture regimes, Newhall, Swap, Soil Taxonomy

Traditional knowledge among Zapotecs of Sierra Madre Del Sur, Oaxaca. Does it represent a base for plant resources management and conservation?

PubMed

Luna-José, Azucena de Lourdes; Aguilar, Beatriz Rendón

2012-07-12

Traditional classification systems represent cognitive processes of human cultures in the world. It synthesizes specific conceptions of nature, as well as cumulative learning, beliefs and customs that are part of a particular human community or society. Traditional knowledge has been analyzed from different viewpoints, one of which corresponds to the analysis of ethnoclassifications. In this work, a brief analysis of the botanical traditional knowledge among Zapotecs of the municipality of San Agustin Loxicha, Oaxaca was conducted. The purposes of this study were: a) to analyze the traditional ecological knowledge of local plant resources through the folk classification of both landscapes and plants and b) to determine the role that this knowledge has played in plant resource management and conservation. The study was developed in five communities of San Agustín Loxicha. From field trips, plant specimens were collected and showed to local people in order to get the Spanish or Zapotec names; through interviews with local people, we obtained names and identified classification categories of plants, vegetation units, and soil types. We found a logic structure in Zapotec plant names, based on linguistic terms, as well as morphological and ecological caracteristics. We followed the classification principles proposed by Berlin [6] in order to build a hierarchical structure of life forms, names and other characteristics mentioned by people. We recorded 757 plant names. Most of them (67%) have an equivalent Zapotec name and the remaining 33% had mixed names with Zapotec and Spanish terms. Plants were categorized as native plants, plants introduced in pre-Hispanic times, or plants introduced later. All of them are grouped in a hierarchical classification, which include life form, generic, specific, and varietal categories. Monotypic and polytypic names are used to further classify plants. This holistic classification system plays an important role for local people in many aspects: it helps to organize and make sense of the diversity, to understand the interrelation among plants-soil-vegetation and to classify their physical space since they relate plants with a particular vegetation unit and a kind of soil. The locals also make a rational use of these elements, because they know which crops can grow in any vegetation unit, or which places are indicated to recollect plants. These aspects are interconnected and could be fundamental for a rational use and management of plant resources.

Comparison of six fire severity classification methods using Montana and Washington wildland fires

Treesearch

Pamela G. Sikkink

2015-01-01

Fire severity classifications are used in the post-fire environment to describe fire effects, such as soil alteration or fuel consumption, on the forest floor. Most of the developed classifications are limited because they address very specific fire effects or post-burn characteristics in the burned environment. However, because fire effects vary so much among soil,...

Soil indigenous knowledge in North Central Namibia

NASA Astrophysics Data System (ADS)

Prudat, Brice; Bloemertz, Lena; Kuhn, Nikolaus J.

2016-04-01

Mapping and classifying soils is part of an important learning process to improve soil management practices, soil quality and increase productivity. In order to assess soil quality improvement related to an ongoing land reform in North-Central Namibia, the characteristics that determine soil quality in the local land use context were determined in this study. To do so, we collated the indigenous soil knowledge in North-Central Namibia where the Ovakwanyama cultivate pearl millet for centuries. Local soil groups are defined mostly based on their productivity potential, which varies depending on the rainfall pattern. The morphological criteria used by the farmers to differentiate the soil groups (colour, consistence) were supported by a conventional analysis of soil physical and chemical properties. Now, they can be used to develop a soil quality assessment toolbox adapted to the regional use. The characteristics of the tool box do not directly indicate soil quality, but refer to local soils groups. The quality of these groups is relatively homogenous at the local scale. Our results show that understanding of indigenous soil knowledge has great potential to improve soil quality assessment with regards to land use. The integration of this knowledge with the conventional soil analysis improves the local meaning of such a "scientific" assessment and thus facilitates dialog between farmers and agronomists, but also scientists working in different regions of the world, but in similar conditions. Overall, the integration of indigenous knowledge in international classification systems (e.g. WRB) as attempted in this study has thus a major potential to improve soil mapping in the local context.

The History of Soil Mapping and Classification in Europe: The role of the European Commission

NASA Astrophysics Data System (ADS)

Montanarella, Luca

2014-05-01

Early systematic soil mapping in Europe dates back to the early times of soil science in the 19th Century and was developed at National scales mostly for taxation purposes. National soil classification systems emerged out of the various scientific communities active at that time in leading countries like Germany, Austria, France, Belgium, United Kingdom and many others. Different scientific communities were leading in the various countries, in some cases stemming from geological sciences, in others as a branch of agricultural sciences. Soil classification for the purpose of ranking soils for their capacity to be agriculturally productive emerged as the main priority, allowing in some countries for very detailed and accurate soil maps at 1:5,000 scale and larger. Detailed mapping was mainly driven by taxation purposes in the early times but evolved in several countries also as a planning and management tool for farms and local administrations. The need for pan-European soil mapping and classification efforts emerged only after World War II in the early 1950's under the auspices of FAO with the aim to compile a common European soil map as a contribution to the global soil mapping efforts of FAO at that time. These efforts evolved over the next decades, with the support of the European Commission, towards the establishment of a permanent network of National soil survey institutions (the European Soil Bureau Network). With the introduction of digital soil mapping technologies, the new European Soil Information System (EUSIS) was established, incorporating data at multiple scales for the EU member states and bordering countries. In more recent years, the formal establishment of the European Soil Data Centre (ESDAC) hosted by the European Commission, together with a formal legal framework for soil mapping and soil classification provided by the INSPIRE directive and the related standardization and harmonization efforts, has led to the operational development of advanced digital soil mapping techniques supporting the contribution of Europe to a common global soil information system under the coordination of the Global Soil Partnership (GSP) of FAO. Further information: http://eusoils.jrc.ec.europa.eu/ References: Mark G Kibblewhite, Ladislav Miko, Luca Montanarella, Legal frameworks for soil protection: current development and technical information requirements, Current Opinion in Environmental Sustainability, Volume 4, Issue 5, November 2012, Pages 573-577. Luca Montanarella, Ronald Vargas, Global governance of soil resources as a necessary condition for sustainable development, Current Opinion in Environmental Sustainability, Volume 4, Issue 5, November 2012, Pages 559-564.

"DEAR ROCK, WHAT'S YOUR DESTINY? Ancient and modern uses of rocks in industry, building and art."

NASA Astrophysics Data System (ADS)

Pennesi, Daniela

2015-04-01

The project is for students of first grade of secondary school. The activity is a game, virtual or real of associations between rock and soil samples with their uses in industry, building and art. The students, alone or in a team, have to form pairs having available various samples of rocks, soils and building materials as bags of cement, tiles.. They have images of colonnades, staircases of famous churches, cave paintings and colors. The project is multidisciplinary. During the activity, the teachers of art and technical education are involved with and the teacher of sciences. The game can be used as an introduction for the rocks' classification. The inquiry in team, is a good way to learn the several uses of mineral resources.

Profile of a city: characterizing and classifying urban soils in the city of Ghent

NASA Astrophysics Data System (ADS)

Delbecque, Nele; Verdoodt, Ann

2017-04-01

Worldwide, urban lands are expanding rapidly. Conversion of agricultural and natural landscapes to urban fabric can strongly influence soil properties through soil sealing, excavation, leveling, contamination, waste disposal and land management. Urban lands, often characterized by intensive use, need to deliver many production, ecological and cultural ecosystem services. To safeguard this natural capital for future generations, an improved understanding of biogeochemical characteristics, processes and functions of urban soils in time and space is essential. Additionally, existing (inter)national soil classification systems, based on the identification of soil genetic horizons, do not always allow a functional classification of urban soils. This research aims (1) to gain insight into urban soils and their properties in the city of Ghent (Belgium), and (2) to develop a procedure to functionally incorporate urban soils into existing (inter)national soil classification systems. Undisturbed soil cores (depth up to 1.25 m) are collected at 15 locations in Ghent with different times since development and land uses. Geotek MSCL-scans are taken to determine magnetic susceptibility and gamma density and to obtain high resolution images. Physico-chemical characterization of the soil cores is performed by means of detailed soil profile descriptions, traditional lab analyses, as well as proximal soil sensing techniques (XRF). The first results of this research will be presented and critically discussed to improve future efforts to characterize, classify and evaluate urban soils and their ecosystem services.

A simulation study of scene confusion factors in sensing soil moisture from orbital radar

NASA Technical Reports Server (NTRS)

Ulaby, F. T. (Principal Investigator); Dobson, M. C.; Moezzi, S.; Roth, F. T.

1983-01-01

Simulated C-band radar imagery for a 124-km by 108-km test site in eastern Kansas is used to classify soil moisture. Simulated radar resolutions are 100 m by 100 m, 1 km by 1km, and 3 km by 3 km. Distributions of actual near-surface soil moisture are established daily for a 23-day accounting period using a water budget model. Within the 23-day period, three orbital radar overpasses are simulated roughly corresponding to generally moist, wet, and dry soil moisture conditions. The radar simulations are performed by a target/sensor interaction model dependent upon a terrain model, land-use classification, and near-surface soil moisture distribution. The accuracy of soil-moisture classification is evaluated for each single-date radar observation and also for multi-date detection of relative soil moisture change. In general, the results for single-date moisture detection show that 70% to 90% of cropland can be correctly classified to within +/- 20% of the true percent of field capacity. For a given radar resolution, the expected classification accuracy is shown to be dependent upon both the general soil moisture condition and also the geographical distribution of land-use and topographic relief. An analysis of cropland, urban, pasture/rangeland, and woodland subregions within the test site indicates that multi-temporal detection of relative soil moisture change is least sensitive to classification error resulting from scene complexity and topographic effects.

Spectroscopic Diagnosis of Arsenic Contamination in Agricultural Soils

PubMed Central

Shi, Tiezhu; Liu, Huizeng; Chen, Yiyun; Fei, Teng; Wang, Junjie; Wu, Guofeng

2017-01-01

This study investigated the abilities of pre-processing, feature selection and machine-learning methods for the spectroscopic diagnosis of soil arsenic contamination. The spectral data were pre-processed by using Savitzky-Golay smoothing, first and second derivatives, multiplicative scatter correction, standard normal variate, and mean centering. Principle component analysis (PCA) and the RELIEF algorithm were used to extract spectral features. Machine-learning methods, including random forests (RF), artificial neural network (ANN), radial basis function- and linear function- based support vector machine (RBF- and LF-SVM) were employed for establishing diagnosis models. The model accuracies were evaluated and compared by using overall accuracies (OAs). The statistical significance of the difference between models was evaluated by using McNemar’s test (Z value). The results showed that the OAs varied with the different combinations of pre-processing, feature selection, and classification methods. Feature selection methods could improve the modeling efficiencies and diagnosis accuracies, and RELIEF often outperformed PCA. The optimal models established by RF (OA = 86%), ANN (OA = 89%), RBF- (OA = 89%) and LF-SVM (OA = 87%) had no statistical difference in diagnosis accuracies (Z < 1.96, p < 0.05). These results indicated that it was feasible to diagnose soil arsenic contamination using reflectance spectroscopy. The appropriate combination of multivariate methods was important to improve diagnosis accuracies. PMID:28471412

Corn and soybean Landsat MSS classification performance as a function of scene characteristics

NASA Technical Reports Server (NTRS)

Batista, G. T.; Hixson, M. M.; Bauer, M. E.

1982-01-01

In order to fully utilize remote sensing to inventory crop production, it is important to identify the factors that affect the accuracy of Landsat classifications. The objective of this study was to investigate the effect of scene characteristics involving crop, soil, and weather variables on the accuracy of Landsat classifications of corn and soybeans. Segments sampling the U.S. Corn Belt were classified using a Gaussian maximum likelihood classifier on multitemporally registered data from two key acquisition periods. Field size had a strong effect on classification accuracy with small fields tending to have low accuracies even when the effect of mixed pixels was eliminated. Other scene characteristics accounting for variability in classification accuracy included proportions of corn and soybeans, crop diversity index, proportion of all field crops, soil drainage, slope, soil order, long-term average soybean yield, maximum yield, relative position of the segment in the Corn Belt, weather, and crop development stage.

Sentiment classification technology based on Markov logic networks

NASA Astrophysics Data System (ADS)

He, Hui; Li, Zhigang; Yao, Chongchong; Zhang, Weizhe

2016-07-01

With diverse online media emerging, there is a growing concern of sentiment classification problem. At present, text sentiment classification mainly utilizes supervised machine learning methods, which feature certain domain dependency. On the basis of Markov logic networks (MLNs), this study proposed a cross-domain multi-task text sentiment classification method rooted in transfer learning. Through many-to-one knowledge transfer, labeled text sentiment classification, knowledge was successfully transferred into other domains, and the precision of the sentiment classification analysis in the text tendency domain was improved. The experimental results revealed the following: (1) the model based on a MLN demonstrated higher precision than the single individual learning plan model. (2) Multi-task transfer learning based on Markov logical networks could acquire more knowledge than self-domain learning. The cross-domain text sentiment classification model could significantly improve the precision and efficiency of text sentiment classification.

A Comparison of Machine Learning Approaches for Corn Yield Estimation

NASA Astrophysics Data System (ADS)

Kim, N.; Lee, Y. W.

2017-12-01

Machine learning is an efficient empirical method for classification and prediction, and it is another approach to crop yield estimation. The objective of this study is to estimate corn yield in the Midwestern United States by employing the machine learning approaches such as the support vector machine (SVM), random forest (RF), and deep neural networks (DNN), and to perform the comprehensive comparison for their results. We constructed the database using satellite images from MODIS, the climate data of PRISM climate group, and GLDAS soil moisture data. In addition, to examine the seasonal sensitivities of corn yields, two period groups were set up: May to September (MJJAS) and July and August (JA). In overall, the DNN showed the highest accuracies in term of the correlation coefficient for the two period groups. The differences between our predictions and USDA yield statistics were about 10-11 %.

Working with Soil - Soil science in the field

NASA Astrophysics Data System (ADS)

Hannam, Jacqueline; Lacelles, Bruce; Owen, Jason; Thompson, Dick; Jones, Bob; Towers, Willie

2015-04-01

Working with Soil is the Professional Competency Scheme developed by the British Society of Soil Science's Professional Practice Committee, formerly the Institute of Professional Soil Scientists. Ten competency documents cover the required qualifications, skills and knowledge for different aspects of applied soil science. The Society is currently engaged in a five year plan to translate the competency documents into a comprehensive set of training courses. Foundation skills in field-based science are covered by three separate training courses - Exposing and describing a soil profile (Course 1), Soil classification (Course 2), and Soil survey techniques (Course 3). Course 1 has run successfully twice a year since 2013. The other two courses are under development and are scheduled to start in 2015. The primary objective of Foundation Skills Course 1 is to develop confidence and familiarity with field soil investigation and description, understanding the soil underfoot and putting soils into a wider landscape context. Delegates excavate a soil profile pit, and describe and sample the exposed soil to standard protocols. Delegates work in teams of 4 or 5 so that an element of shared learning is part of the process. This has been a very positive aspect of the courses we have run to date. The course has attracted professionals from agricultural and environmental consultancies but is also very popular with research students and has formed a part of an Advanced Training Programme in Soil Science for postgraduates. As there is only one soil science degree course remaining in the UK, many students on their admission do not have a background in field-based pedology and lack an understanding of soil in the context of landscape scale soil functions. Feedback to date has been very positive.

Selected Aspects of Soil Science History in the USA - Prehistory to the 1970s

NASA Astrophysics Data System (ADS)

Brevik, Eric C.; Fenton, Thomas E.; Homburg, Jeffrey A.

2017-04-01

Interest in understanding America's soils originated in prehistory with Native Americans. Following European settlement, notable individuals such as Thomas Jefferson and Lewis and Clark made observations of soil resources. Moving into the 1800s, state geological surveys became involved in soil work and E.W. Hilgard started to formulate ideas similar to those that would eventually lead to V.V. Dokuchaev being recognized as the father of modern soil science. However, Hilgard's advanced ideas on soil genesis were not accepted by the wider American soil science community at the time. Moving into the 1900s, the National Cooperative Soil Survey, the first nationally organized detailed soil survey in the world, was founded under the direction of M. Whitney. Initial soil classification ideas were heavily based in geology, but over time Russian ideas of soil genesis and classification moved into the American soil science community, mainly due to the influence of C.F. Marbut. Early American efforts in scientific study of soil erosion and soil fertility were also initiated in the 1910s and university programs to educate soil scientists started. Soil erosion studies took on high priority in the 1930s as the USA was impacted by the Dust Bowl. Soil Taxonomy, one of the most widely utilized soil classification systems in the world, was developed from the 1950s through the 1970s under the guidance of G.D. Smith and with administrative support from C.E. Kellogg. American soil scientists, such as H. Jenny, R.W. Simonson, D.L. Johnson, and D. Watson-Stegner, developed influential models of soil genesis during the 20th Century, and the use of soil information expanded beyond agriculture to include issues such as land-use planning, soil geomorphology, and interactions between soils and human health.

Characteristic variations in reflectance of surface soils

NASA Technical Reports Server (NTRS)

Stoner, E. R.; Baumgardner, M. F. (Principal Investigator)

1982-01-01

Surface soil samples from a wide range of naturally occurring soils were obtained for the purpose of studying the characteristic variations in soil reflectance as these variations relate to other soil properties and soil classification. A total 485 soil samples from the U.S. and Brazil representing 30 suborders of the 10 orders of 'Soil Taxonomy' was examined. The spectral bidirectional reflectance factor was measured on uniformly moist soils over the 0.52 to 2.32 micron wavelength range with a spectroradiometer adapted for indoor use. Five distinct soil spectral reflectance curve forms were identified according to curve shape, the presence or absence of absorption bands, and the predominance of soil organic matter and iron oxide composition. These curve forms were further characterized according to generically homogeneous soil properties in a manner similar to the subdivisions at the suborder level of 'Soil Taxonomy'. Results indicate that spectroradiometric measurements of soil spectral bidirectional reflectance factor can be used to characterize soil reflectance in terms that are meaningful to soil classification, genesis, and survey.

USCS and the USDA Soil Classification System: Development of a Mapping Scheme

DTIC Science & Technology

2015-03-01

important to human daily living. A variety of disciplines (geology, agriculture, engineering, etc.) require a sys- tematic categorization of soil, detailing...it is often important to also con- sider parameters that indicate soil strength. Two important properties used for engineering-related problems are...that many textural clas- sification systems were developed to meet specifics needs. In agriculture, textural classification is used to determine crop

Training strategy for convolutional neural networks in pedestrian gender classification

NASA Astrophysics Data System (ADS)

Ng, Choon-Boon; Tay, Yong-Haur; Goi, Bok-Min

2017-06-01

In this work, we studied a strategy for training a convolutional neural network in pedestrian gender classification with limited amount of labeled training data. Unsupervised learning by k-means clustering on pedestrian images was used to learn the filters to initialize the first layer of the network. As a form of pre-training, supervised learning for the related task of pedestrian classification was performed. Finally, the network was fine-tuned for gender classification. We found that this strategy improved the network's generalization ability in gender classification, achieving better test results when compared to random weights initialization and slightly more beneficial than merely initializing the first layer filters by unsupervised learning. This shows that unsupervised learning followed by pre-training with pedestrian images is an effective strategy to learn useful features for pedestrian gender classification.

Human Factors Engineering. Student Supplement,

DTIC Science & Technology

1981-08-01

a job TASK TAXONOMY A classification scheme for the different levels of activities in a system, i.e., job - task - sub-task, etc. TASK-AN~ALYSIS...with the classification of learning objectives by learning category so as to identify learningPhas III guidelines necessary for optimum learning to...correct. .4... .the sequencing of all dependent tasks. .1.. .the classification of learning objectives by learning category and the Identification of

Comparative analysis of tree classification models for detecting fusarium oxysporum f. sp cubense (TR4) based on multi soil sensor parameters

NASA Astrophysics Data System (ADS)

Estuar, Maria Regina Justina; Victorino, John Noel; Coronel, Andrei; Co, Jerelyn; Tiausas, Francis; Señires, Chiara Veronica

2017-09-01

Use of wireless sensor networks and smartphone integration design to monitor environmental parameters surrounding plantations is made possible because of readily available and affordable sensors. Providing low cost monitoring devices would be beneficial, especially to small farm owners, in a developing country like the Philippines, where agriculture covers a significant amount of the labor market. This study discusses the integration of wireless soil sensor devices and smartphones to create an application that will use multidimensional analysis to detect the presence or absence of plant disease. Specifically, soil sensors are designed to collect soil quality parameters in a sink node from which the smartphone collects data from via Bluetooth. Given these, there is a need to develop a classification model on the mobile phone that will report infection status of a soil. Though tree classification is the most appropriate approach for continuous parameter-based datasets, there is a need to determine whether tree models will result to coherent results or not. Soil sensor data that resides on the phone is modeled using several variations of decision tree, namely: decision tree (DT), best-fit (BF) decision tree, functional tree (FT), Naive Bayes (NB) decision tree, J48, J48graft and LAD tree, where decision tree approaches the problem by considering all sensor nodes as one. Results show that there are significant differences among soil sensor parameters indicating that there are variances in scores between the infected and uninfected sites. Furthermore, analysis of variance in accuracy, recall, precision and F1 measure scores from tree classification models homogeneity among NBTree, J48graft and J48 tree classification models.

Learning classification models with soft-label information.

PubMed

Nguyen, Quang; Valizadegan, Hamed; Hauskrecht, Milos

2014-01-01

Learning of classification models in medicine often relies on data labeled by a human expert. Since labeling of clinical data may be time-consuming, finding ways of alleviating the labeling costs is critical for our ability to automatically learn such models. In this paper we propose a new machine learning approach that is able to learn improved binary classification models more efficiently by refining the binary class information in the training phase with soft labels that reflect how strongly the human expert feels about the original class labels. Two types of methods that can learn improved binary classification models from soft labels are proposed. The first relies on probabilistic/numeric labels, the other on ordinal categorical labels. We study and demonstrate the benefits of these methods for learning an alerting model for heparin induced thrombocytopenia. The experiments are conducted on the data of 377 patient instances labeled by three different human experts. The methods are compared using the area under the receiver operating characteristic curve (AUC) score. Our AUC results show that the new approach is capable of learning classification models more efficiently compared to traditional learning methods. The improvement in AUC is most remarkable when the number of examples we learn from is small. A new classification learning framework that lets us learn from auxiliary soft-label information provided by a human expert is a promising new direction for learning classification models from expert labels, reducing the time and cost needed to label data.

Taxonomic classification of world map units in crop producing areas of Argentina and Brazil with representative US soil series and major land resource areas in which they occur

NASA Technical Reports Server (NTRS)

Huckle, H. F. (Principal Investigator)

1980-01-01

The most probable current U.S. taxonomic classification of the soils estimated to dominate world soil map units (WSM)) in selected crop producing states of Argentina and Brazil are presented. Representative U.S. soil series the units are given. The map units occurring in each state are listed with areal extent and major U.S. land resource areas in which similar soils most probably occur. Soil series sampled in LARS Technical Report 111579 and major land resource areas in which they occur with corresponding similar WSM units at the taxonomic subgroup levels are given.

Developments and departures in the philosophy of soil science

USDA-ARS?s Scientific Manuscript database

Traditional soil science curriculums provide comprehensive instruction on soil properties, soil classification, and the physical, chemical, and biological processes that occur in soils. This reductionist perspective is sometimes balanced with a more holistic perspective that focuses on soils as natu...

Classification versus inference learning contrasted with real-world categories.

PubMed

Jones, Erin L; Ross, Brian H

2011-07-01

Categories are learned and used in a variety of ways, but the research focus has been on classification learning. Recent work contrasting classification with inference learning of categories found important later differences in category performance. However, theoretical accounts differ on whether this is due to an inherent difference between the tasks or to the implementation decisions. The inherent-difference explanation argues that inference learners focus on the internal structure of the categories--what each category is like--while classification learners focus on diagnostic information to predict category membership. In two experiments, using real-world categories and controlling for earlier methodological differences, inference learners learned more about what each category was like than did classification learners, as evidenced by higher performance on a novel classification test. These results suggest that there is an inherent difference between learning new categories by classifying an item versus inferring a feature.

Support-vector-machine tree-based domain knowledge learning toward automated sports video classification

NASA Astrophysics Data System (ADS)

Xiao, Guoqiang; Jiang, Yang; Song, Gang; Jiang, Jianmin

2010-12-01

We propose a support-vector-machine (SVM) tree to hierarchically learn from domain knowledge represented by low-level features toward automatic classification of sports videos. The proposed SVM tree adopts a binary tree structure to exploit the nature of SVM's binary classification, where each internal node is a single SVM learning unit, and each external node represents the classified output type. Such a SVM tree presents a number of advantages, which include: 1. low computing cost; 2. integrated learning and classification while preserving individual SVM's learning strength; and 3. flexibility in both structure and learning modules, where different numbers of nodes and features can be added to address specific learning requirements, and various learning models can be added as individual nodes, such as neural networks, AdaBoost, hidden Markov models, dynamic Bayesian networks, etc. Experiments support that the proposed SVM tree achieves good performances in sports video classifications.

Site classification for northern forest species

Treesearch

Willard H. Carmean

1977-01-01

Summarizes the extensive literature for northern forest species covering site index curves, site index species comparisons, growth intercepts, soil-site studies, plant indicators, physiographic site classifications, and soil survey studies. The advantages and disadvantages of each are discussed, and suggestions are made for future research using each of these methods....

7 CFR 658.5 - Criteria.

Code of Federal Regulations, 2011 CFR

2011-01-01

... information from several sources including national cooperative soil surveys or other acceptable soil surveys, NRCS field office technical guides, soil potential ratings or soil productivity ratings, land capability classifications, and important farmland determinations. Based on this information, groups of soils...

7 CFR 658.5 - Criteria.

Code of Federal Regulations, 2010 CFR

2010-01-01

... information from several sources including national cooperative soil surveys or other acceptable soil surveys, NRCS field office technical guides, soil potential ratings or soil productivity ratings, land capability classifications, and important farmland determinations. Based on this information, groups of soils...

Feasibility of Active Machine Learning for Multiclass Compound Classification.

PubMed

Lang, Tobias; Flachsenberg, Florian; von Luxburg, Ulrike; Rarey, Matthias

2016-01-25

A common task in the hit-to-lead process is classifying sets of compounds into multiple, usually structural classes, which build the groundwork for subsequent SAR studies. Machine learning techniques can be used to automate this process by learning classification models from training compounds of each class. Gathering class information for compounds can be cost-intensive as the required data needs to be provided by human experts or experiments. This paper studies whether active machine learning can be used to reduce the required number of training compounds. Active learning is a machine learning method which processes class label data in an iterative fashion. It has gained much attention in a broad range of application areas. In this paper, an active learning method for multiclass compound classification is proposed. This method selects informative training compounds so as to optimally support the learning progress. The combination with human feedback leads to a semiautomated interactive multiclass classification procedure. This method was investigated empirically on 15 compound classification tasks containing 86-2870 compounds in 3-38 classes. The empirical results show that active learning can solve these classification tasks using 10-80% of the data which would be necessary for standard learning techniques.

Multi-temporal Land Use Mapping of Coastal Wetlands Area using Machine Learning in Google Earth Engine

NASA Astrophysics Data System (ADS)

Farda, N. M.

2017-12-01

Coastal wetlands provide ecosystem services essential to people and the environment. Changes in coastal wetlands, especially on land use, are important to monitor by utilizing multi-temporal imagery. The Google Earth Engine (GEE) provides many machine learning algorithms (10 algorithms) that are very useful for extracting land use from imagery. The research objective is to explore machine learning in Google Earth Engine and its accuracy for multi-temporal land use mapping of coastal wetland area. Landsat 3 MSS (1978), Landsat 5 TM (1991), Landsat 7 ETM+ (2001), and Landsat 8 OLI (2014) images located in Segara Anakan lagoon are selected to represent multi temporal images. The input for machine learning are visible and near infrared bands, PCA band, invers PCA bands, bare soil index, vegetation index, wetness index, elevation from ASTER GDEM, and GLCM (Harralick) texture, and also polygon samples in 140 locations. There are 10 machine learning algorithms applied to extract coastal wetlands land use from Landsat imagery. The algorithms are Fast Naive Bayes, CART (Classification and Regression Tree), Random Forests, GMO Max Entropy, Perceptron (Multi Class Perceptron), Winnow, Voting SVM, Margin SVM, Pegasos (Primal Estimated sub-GrAdient SOlver for Svm), IKPamir (Intersection Kernel Passive Aggressive Method for Information Retrieval, SVM). Machine learning in Google Earth Engine are very helpful in multi-temporal land use mapping, the highest accuracy for land use mapping of coastal wetland is CART with 96.98 % Overall Accuracy using K-Fold Cross Validation (K = 10). GEE is particularly useful for multi-temporal land use mapping with ready used image and classification algorithms, and also very challenging for other applications.

Relationship between the erosion properties of soils and other parameters

USDA-ARS?s Scientific Manuscript database

Soil parameters are essential for erosion process prediction and ultimately improved model development, especially as they relate to dam and levee failure. Soil parameters including soil texture and structure, soil classification, soil compaction, moisture content, and degree of saturation can play...

Polarimetric SAR image classification based on discriminative dictionary learning model

NASA Astrophysics Data System (ADS)

Sang, Cheng Wei; Sun, Hong

2018-03-01

Polarimetric SAR (PolSAR) image classification is one of the important applications of PolSAR remote sensing. It is a difficult high-dimension nonlinear mapping problem, the sparse representations based on learning overcomplete dictionary have shown great potential to solve such problem. The overcomplete dictionary plays an important role in PolSAR image classification, however for PolSAR image complex scenes, features shared by different classes will weaken the discrimination of learned dictionary, so as to degrade classification performance. In this paper, we propose a novel overcomplete dictionary learning model to enhance the discrimination of dictionary. The learned overcomplete dictionary by the proposed model is more discriminative and very suitable for PolSAR classification.

Integrating Interview Methodology to Analyze Inter-Institutional Comparisons of Service-Learning within the Carnegie Community Engagement Classification Framework

ERIC Educational Resources Information Center

Plante, Jarrad D.; Cox, Thomas D.

2016-01-01

Service-learning has a longstanding history in higher education in and includes three main tenets: academic learning, meaningful community service, and civic learning. The Carnegie Foundation for the Advancement of Teaching created an elective classification system called the Carnegie Community Engagement Classification for higher education…

The Costs of Supervised Classification: The Effect of Learning Task on Conceptual Flexibility

ERIC Educational Resources Information Center

Hoffman, Aaron B.; Rehder, Bob

2010-01-01

Research has shown that learning a concept via standard supervised classification leads to a focus on diagnostic features, whereas learning by inferring missing features promotes the acquisition of within-category information. Accordingly, we predicted that classification learning would produce a deficit in people's ability to draw "novel…

Learning about the internal structure of categories through classification and feature inference.

PubMed

Jee, Benjamin D; Wiley, Jennifer

2014-01-01

Previous research on category learning has found that classification tasks produce representations that are skewed toward diagnostic feature dimensions, whereas feature inference tasks lead to richer representations of within-category structure. Yet, prior studies often measure category knowledge through tasks that involve identifying only the typical features of a category. This neglects an important aspect of a category's internal structure: how typical and atypical features are distributed within a category. The present experiments tested the hypothesis that inference learning results in richer knowledge of internal category structure than classification learning. We introduced several new measures to probe learners' representations of within-category structure. Experiment 1 found that participants in the inference condition learned and used a wider range of feature dimensions than classification learners. Classification learners, however, were more sensitive to the presence of atypical features within categories. Experiment 2 provided converging evidence that classification learners were more likely to incorporate atypical features into their representations. Inference learners were less likely to encode atypical category features, even in a "partial inference" condition that focused learners' attention on the feature dimensions relevant to classification. Overall, these results are contrary to the hypothesis that inference learning produces superior knowledge of within-category structure. Although inference learning promoted representations that included a broad range of category-typical features, classification learning promoted greater sensitivity to the distribution of typical and atypical features within categories.

DecoFungi: a web application for automatic characterisation of dye decolorisation in fungal strains.

PubMed

Domínguez, César; Heras, Jónathan; Mata, Eloy; Pascual, Vico

2018-02-27

Fungi have diverse biotechnological applications in, among others, agriculture, bioenergy generation, or remediation of polluted soil and water. In this context, culture media based on color change in response to degradation of dyes are particularly relevant; but measuring dye decolorisation of fungal strains mainly relies on a visual and semiquantitative classification of color intensity changes. Such a classification is a subjective, time-consuming and difficult to reproduce process. DecoFungi is the first, at least up to the best of our knowledge, application to automatically characterise dye decolorisation level of fungal strains from images of inoculated plates. In order to deal with this task, DecoFungi employs a deep-learning model, accessible through a user-friendly web interface, with an accuracy of 96.5%. DecoFungi is an easy to use system for characterising dye decolorisation level of fungal strains from images of inoculated plates.

Classification and Use of Natural and Anthropogenic Soils by Indigenous Communities of the Upper Amazon Region of Colombia.

PubMed

Peña-Venegas, C P; Stomph, T J; Verschoor, G; Echeverri, J A; Struik, P C

Outsiders often oversimplify Amazon soil use by assuming that abundantly available natural soils are poorly suited to agriculture and that sporadic anthropogenic soils are agriculturally productive. Local perceptions about the potentials and limitations of soils probably differ, but information on these perceptions is scarce. We therefore examined how four indigenous communities in the Middle Caquetá River region in the Colombian Amazon classify and use natural and anthropogenic soils. The study was framed in ethnopedology: local classifications, preferences, rankings, and soil uses were recorded through interviews and field observations. These communities recognized nine soils varying in suitability for agriculture. They identified anthropogenic soils as most suitable for agriculture, but only one group used them predominantly for their swiddens. As these communities did not perceive soil nutrient status as limiting, they did not base crop-site selection on soil fertility or on the interplay between soil quality and performance of manioc genetic resources.

Methods for Tier 1 Modeling within the Training Range Environmental Evaluation and Characterization System

DTIC Science & Technology

2009-08-01

properties, part b. USLE K-Factor by Organic Matter Content Soil -Texture Classification Dry Bulk Density, g/cm3 Field Capacity, % Available...Universal Soil Loss Equation ( USLE ) can be used to estimate annual average sheet and rill erosion, A (tons/acre-yr), from the equation A R K L S...erodibility factors, K, for various soil classifications and percent organic matter content ( USLE Fact Sheet 2008). Textural Class Average Less than 2

EXTENDING AQUATIC CLASSIFICATION TO THE LANDSCAPE SCALE HYDROLOGY-BASED STRATEGIES

EPA Science Inventory

Aquatic classification of single water bodies (lakes, wetlands, estuaries) is often based on geologic origin, while stream classification has relied on multiple factors related to landform, geomorphology, and soils. We have developed an approach to aquatic classification based o...

Guidance for Determining the Acceptability of Environmental Fate Studies Conducted with Foreign Soils

EPA Pesticide Factsheets

International and U.S. soil taxonomical classification systems, distribution of soil orders in the United States, specific criteria to help scientists determine when foreign soils are representative of U.S. soils at intended pesticide use sites.

Landcover Classification Using Deep Fully Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Wang, J.; Li, X.; Zhou, S.; Tang, J.

2017-12-01

Land cover classification has always been an essential application in remote sensing. Certain image features are needed for land cover classification whether it is based on pixel or object-based methods. Different from other machine learning methods, deep learning model not only extracts useful information from multiple bands/attributes, but also learns spatial characteristics. In recent years, deep learning methods have been developed rapidly and widely applied in image recognition, semantic understanding, and other application domains. However, there are limited studies applying deep learning methods in land cover classification. In this research, we used fully convolutional networks (FCN) as the deep learning model to classify land covers. The National Land Cover Database (NLCD) within the state of Kansas was used as training dataset and Landsat images were classified using the trained FCN model. We also applied an image segmentation method to improve the original results from the FCN model. In addition, the pros and cons between deep learning and several machine learning methods were compared and explored. Our research indicates: (1) FCN is an effective classification model with an overall accuracy of 75%; (2) image segmentation improves the classification results with better match of spatial patterns; (3) FCN has an excellent ability of learning which can attains higher accuracy and better spatial patterns compared with several machine learning methods.

Land cover heterogeneity and soil respiration in a west Greenland tundra landscape

NASA Astrophysics Data System (ADS)

Bradley-Cook, J. I.; Burzynski, A.; Hammond, C. R.; Virginia, R. A.

2011-12-01

Multiple direct and indirect pathways underlie the association between land cover classification, temperature and soil respiration. Temperature is a main control of the biological processes that constitute soil respiration, yet the effect of changing atmospheric temperatures on soil carbon flux is unresolved. This study examines associations amongst land cover, soil carbon characteristics, soil respiration, and temperature in an Arctic tundra landscape in western Greenland. We used a 1.34 meter resolution multi-spectral WorldView2 satellite image to conduct an unsupervised multi-staged ISODATA classification to characterize land cover heterogeneity. The four band image was taken on July 10th, 2010, and captures an 18 km by 15 km area in the vicinity of Kangerlussuaq. The four major terrestrial land cover classes identified were: shrub-dominated, graminoid-dominated, mixed vegetation, and bare soil. The bare soil class was comprised of patches where surface soil has been deflated by wind and ridge-top fellfield. We hypothesize that soil respiration and soil carbon storage are associated with land cover classification and temperature. We set up a hierarchical field sampling design to directly observe spatial variation between and within land cover classes along a 20 km temperature gradient extending west from Russell Glacier on the margin of the Greenland Ice Sheet. We used the land cover classification map and ground verification to select nine sites, each containing patches of the four land cover classes. Within each patch we collected soil samples from a 50 cm pit, quantified vegetation, measured active layer depth and determined landscape characteristics. From a subset of field sites we collected additional 10 cm surface soil samples to estimate soil heterogeneity within patches and measured soil respiration using a LiCor 8100 Infrared Gas Analyzer. Soil respiration rates varied with land cover classes, with values ranging from 0.2 mg C/m^2/hr in the bare soil class to over 5 mg C/m^2/hr in the graminoid-dominated class. These findings suggest that shifts in land cover vegetation types, especially soil and vegetation loss (e.g. from wind deflation), can alter landscape soil respiration. We relate soil respiration measurements to soil, vegetation, and permafrost characteristics to understand how ecosystem properties and processes vary at the landscape scale. A long-term goal of this research is to develop a spatially explicit model of soil organic matter, soil respiration, and temperature sensitivity of soil carbon dynamics for a western Greenland permafrost tundra ecosystems.

History of Soil Survey and Evolution of the Brazilian Soil Classification System - SiBCS

NASA Astrophysics Data System (ADS)

Cunha dos Anjos, Lúcia Helena; Csekö Nolasco de Carvalho, Claudia; Homem Antunes, Mauro Antonio; Muggler, Cristine Carole

2014-05-01

In Brazil soil surveys started around 1940 and the first map with soil information of São Paulo State was published in 1943. The Committee of Soils of the National Service for Agronomic Research was created in 1947 by the Agriculture Ministry and became an historical landmark for soil survey in Brazil. In 1953, the National Program of soil survey was approved and the first soil map and report of Rio de Janeiro State was released in 1958, followed by São Paulo State in 1960. This is also the origin of Embrapa Soil Research institution. Other milestones were the soil surveys published by the Agronomic Institute of Campinas (IAC) and the natural resources studies published within the RADAMBRASIL Project, initially planned for the Amazon region and later covering the whole country. Many soil studies followed and a comprehensive knowledge of tropical soils was achieved resulting in successful technologies for agriculture production, in lands considered by many as of "low fertility and acid soils with limited or no agricultural potential". However, detailed soil surveys are still lacking; only 5% of the country soils are mapped in 1:25.000 scales, and 15-20% in 1:100.000. In the first soil survey reports of Rio de Janeiro (1958) and São Paulo (1960), soil classes were defined according to Baldwin, Kellog & Thorp (Yearbook of Agriculture for 1938), and Thorp & Smith (Soil Science, 67, 1949) publications. It was already clear that the existing classification systems were not adequate to represent the highly weathered tropical soils of the large old landscapes in the cerrado (savanna like) region, or the soils formed on recent hydromorphic conditions at the Amazon Basin and Pantanal region. A national classification system to embody the country's large territory and environmental variation from tropical to subtropical and semiarid conditions, as well as the diversity of soil forming processes in old and new landscapes had to be developed. In 1964, the first attempt of a national soil classification was presented by Marcelo Camargo (Embrapa Soils) and Jacob Bennema (FAO adviser). When Soil Taxonomy was first published in 1975, a field workshop was held in Brazil, and the system was not accepted by the country scientists; one main reason was the usage of climate as a main attribute for suborders. In 1978, the first national soil field correlation meeting was held with the goal of developing the national system, giving origin to the Brazilian Soil Classification System (SiBCS). In 1980, a working group was created by Embrapa Soils and other institutes resulting in four approximations of the system. In 1999, the first edition of the SiBCS was released, followed by a second edition in 2006 and the third in 2013. The SiBCS is a hierarchic system, based on morphogenetic soil attributes, with six categorical levels: order, suborder, great group, subgroup, family, and series. It has 13 soil orders, and it is structured as a key down to subgroup level. Many soil attributes are based on concepts adopted by the Soil Taxonomy (United States) and by the World Reference Base for Soil Resources (WRB - FAO). The development of the SiBCS is supervised by a national executive committee, and information is available at http://www.cnps.embrapa.br/sibcs (in Portuguese).

Regularised extreme learning machine with misclassification cost and rejection cost for gene expression data classification.

PubMed

Lu, Huijuan; Wei, Shasha; Zhou, Zili; Miao, Yanzi; Lu, Yi

2015-01-01

The main purpose of traditional classification algorithms on bioinformatics application is to acquire better classification accuracy. However, these algorithms cannot meet the requirement that minimises the average misclassification cost. In this paper, a new algorithm of cost-sensitive regularised extreme learning machine (CS-RELM) was proposed by using probability estimation and misclassification cost to reconstruct the classification results. By improving the classification accuracy of a group of small sample which higher misclassification cost, the new CS-RELM can minimise the classification cost. The 'rejection cost' was integrated into CS-RELM algorithm to further reduce the average misclassification cost. By using Colon Tumour dataset and SRBCT (Small Round Blue Cells Tumour) dataset, CS-RELM was compared with other cost-sensitive algorithms such as extreme learning machine (ELM), cost-sensitive extreme learning machine, regularised extreme learning machine, cost-sensitive support vector machine (SVM). The results of experiments show that CS-RELM with embedded rejection cost could reduce the average cost of misclassification and made more credible classification decision than others.

Comparison of different landform classification methods for digital landform and soil mapping of the Iranian loess plateau

NASA Astrophysics Data System (ADS)

Hoffmeister, Dirk; Kramm, Tanja; Curdt, Constanze; Maleki, Sedigheh; Khormali, Farhad; Kehl, Martin

2016-04-01

The Iranian loess plateau is covered by loess deposits, up to 70 m thick. Tectonic uplift triggered deep erosion and valley incision into the loess and underlying marine deposits. Soil development strongly relates to the aspect of these incised slopes, because on northern slopes vegetation protects the soil surface against erosion and facilitates formation and preservation of a Cambisol, whereas on south-facing slopes soils were probably eroded and weakly developed Entisols formed. While the whole area is intensively stocked with sheep and goat, rain-fed cropping of winter wheat is practiced on the valley floors. Most time of the year, the soil surface is unprotected against rainfall, which is one of the factors promoting soil erosion and serious flooding. However, little information is available on soil distribution, plant cover and the geomorphological evolution of the plateau, as well as on potentials and problems in land use. Thus, digital landform and soil mapping is needed. As a requirement of digital landform and soil mapping, four different landform classification methods were compared and evaluated. These geomorphometric classifications were run on two different scales. On the whole area an ASTER GDEM and SRTM dataset (30 m pixel resolution) was used. Likewise, two high-resolution digital elevation models were derived from Pléiades satellite stereo-imagery (< 1m pixel resolution, 10 by 10 km). The high-resolution information of this dataset was aggregated to datasets of 5 and 10 m scale. The applied classification methods are the Geomorphons approach, an object-based image approach, the topographical position index and a mainly slope based approach. The accuracy of the classification was checked with a location related image dataset obtained in a field survey (n ~ 150) in September 2015. The accuracy of the DEMs was compared to measured DGPS trenches and map-based elevation data. The overall derived accuracy of the landform classification based on the high-resolution DEM with a resolution of 5 m is approximately 70% and on a 10 m resolution >58%. For the 30 m resolution datasets is the achieved accuracy approximately 40%, as several small scale features are not recognizable in this resolution. Thus, for an accurate differentiation between different important landform types, high-resolution datasets are necessary for this strongly shaped area. One major problem of this approach are the different classes derived by each method and the various class annotations. The result of this evaluation will be regarded for the derivation of landform and soil maps.

Joint Feature Selection and Classification for Multilabel Learning.

PubMed

Huang, Jun; Li, Guorong; Huang, Qingming; Wu, Xindong

2018-03-01

Multilabel learning deals with examples having multiple class labels simultaneously. It has been applied to a variety of applications, such as text categorization and image annotation. A large number of algorithms have been proposed for multilabel learning, most of which concentrate on multilabel classification problems and only a few of them are feature selection algorithms. Current multilabel classification models are mainly built on a single data representation composed of all the features which are shared by all the class labels. Since each class label might be decided by some specific features of its own, and the problems of classification and feature selection are often addressed independently, in this paper, we propose a novel method which can perform joint feature selection and classification for multilabel learning, named JFSC. Different from many existing methods, JFSC learns both shared features and label-specific features by considering pairwise label correlations, and builds the multilabel classifier on the learned low-dimensional data representations simultaneously. A comparative study with state-of-the-art approaches manifests a competitive performance of our proposed method both in classification and feature selection for multilabel learning.

Unsupervised feature learning for autonomous rock image classification

NASA Astrophysics Data System (ADS)

Shu, Lei; McIsaac, Kenneth; Osinski, Gordon R.; Francis, Raymond

2017-09-01

Autonomous rock image classification can enhance the capability of robots for geological detection and enlarge the scientific returns, both in investigation on Earth and planetary surface exploration on Mars. Since rock textural images are usually inhomogeneous and manually hand-crafting features is not always reliable, we propose an unsupervised feature learning method to autonomously learn the feature representation for rock images. In our tests, rock image classification using the learned features shows that the learned features can outperform manually selected features. Self-taught learning is also proposed to learn the feature representation from a large database of unlabelled rock images of mixed class. The learned features can then be used repeatedly for classification of any subclass. This takes advantage of the large dataset of unlabelled rock images and learns a general feature representation for many kinds of rocks. We show experimental results supporting the feasibility of self-taught learning on rock images.

Estimating Soil Organic Carbon Stocks and Spatial Patterns with Statistical and GIS-Based Methods

PubMed Central

Zhi, Junjun; Jing, Changwei; Lin, Shengpan; Zhang, Cao; Liu, Qiankun; DeGloria, Stephen D.; Wu, Jiaping

2014-01-01

Accurately quantifying soil organic carbon (SOC) is considered fundamental to studying soil quality, modeling the global carbon cycle, and assessing global climate change. This study evaluated the uncertainties caused by up-scaling of soil properties from the county scale to the provincial scale and from lower-level classification of Soil Species to Soil Group, using four methods: the mean, median, Soil Profile Statistics (SPS), and pedological professional knowledge based (PKB) methods. For the SPS method, SOC stock is calculated at the county scale by multiplying the mean SOC density value of each soil type in a county by its corresponding area. For the mean or median method, SOC density value of each soil type is calculated using provincial arithmetic mean or median. For the PKB method, SOC density value of each soil type is calculated at the county scale considering soil parent materials and spatial locations of all soil profiles. A newly constructed 1∶50,000 soil survey geographic database of Zhejiang Province, China, was used for evaluation. Results indicated that with soil classification levels up-scaling from Soil Species to Soil Group, the variation of estimated SOC stocks among different soil classification levels was obviously lower than that among different methods. The difference in the estimated SOC stocks among the four methods was lowest at the Soil Species level. The differences in SOC stocks among the mean, median, and PKB methods for different Soil Groups resulted from the differences in the procedure of aggregating soil profile properties to represent the attributes of one soil type. Compared with the other three estimation methods (i.e., the SPS, mean and median methods), the PKB method holds significant promise for characterizing spatial differences in SOC distribution because spatial locations of all soil profiles are considered during the aggregation procedure. PMID:24840890

The Exploring Nature of Definitions and Classifications of Language Learning Strategies (LLSs) in the Current Studies of Second/Foreign Language Learning

ERIC Educational Resources Information Center

Fazeli, Seyed Hossein

2011-01-01

This study aims to explore the nature of definitions and classifications of Language Learning Strategies (LLSs) in the current studies of second/foreign language learning in order to show the current problems regarding such definitions and classifications. The present study shows that there is not a universal agreeable definition and…

Feature Inference Learning and Eyetracking

ERIC Educational Resources Information Center

Rehder, Bob; Colner, Robert M.; Hoffman, Aaron B.

2009-01-01

Besides traditional supervised classification learning, people can learn categories by inferring the missing features of category members. It has been proposed that feature inference learning promotes learning a category's internal structure (e.g., its typical features and interfeature correlations) whereas classification promotes the learning of…

From learning taxonomies to phylogenetic learning: integration of 16S rRNA gene data into FAME-based bacterial classification.

PubMed

Slabbinck, Bram; Waegeman, Willem; Dawyndt, Peter; De Vos, Paul; De Baets, Bernard

2010-01-30

Machine learning techniques have shown to improve bacterial species classification based on fatty acid methyl ester (FAME) data. Nonetheless, FAME analysis has a limited resolution for discrimination of bacteria at the species level. In this paper, we approach the species classification problem from a taxonomic point of view. Such a taxonomy or tree is typically obtained by applying clustering algorithms on FAME data or on 16S rRNA gene data. The knowledge gained from the tree can then be used to evaluate FAME-based classifiers, resulting in a novel framework for bacterial species classification. In view of learning in a taxonomic framework, we consider two types of trees. First, a FAME tree is constructed with a supervised divisive clustering algorithm. Subsequently, based on 16S rRNA gene sequence analysis, phylogenetic trees are inferred by the NJ and UPGMA methods. In this second approach, the species classification problem is based on the combination of two different types of data. Herein, 16S rRNA gene sequence data is used for phylogenetic tree inference and the corresponding binary tree splits are learned based on FAME data. We call this learning approach 'phylogenetic learning'. Supervised Random Forest models are developed to train the classification tasks in a stratified cross-validation setting. In this way, better classification results are obtained for species that are typically hard to distinguish by a single or flat multi-class classification model. FAME-based bacterial species classification is successfully evaluated in a taxonomic framework. Although the proposed approach does not improve the overall accuracy compared to flat multi-class classification, it has some distinct advantages. First, it has better capabilities for distinguishing species on which flat multi-class classification fails. Secondly, the hierarchical classification structure allows to easily evaluate and visualize the resolution of FAME data for the discrimination of bacterial species. Summarized, by phylogenetic learning we are able to situate and evaluate FAME-based bacterial species classification in a more informative context.

From learning taxonomies to phylogenetic learning: Integration of 16S rRNA gene data into FAME-based bacterial classification

PubMed Central

2010-01-01

Background Machine learning techniques have shown to improve bacterial species classification based on fatty acid methyl ester (FAME) data. Nonetheless, FAME analysis has a limited resolution for discrimination of bacteria at the species level. In this paper, we approach the species classification problem from a taxonomic point of view. Such a taxonomy or tree is typically obtained by applying clustering algorithms on FAME data or on 16S rRNA gene data. The knowledge gained from the tree can then be used to evaluate FAME-based classifiers, resulting in a novel framework for bacterial species classification. Results In view of learning in a taxonomic framework, we consider two types of trees. First, a FAME tree is constructed with a supervised divisive clustering algorithm. Subsequently, based on 16S rRNA gene sequence analysis, phylogenetic trees are inferred by the NJ and UPGMA methods. In this second approach, the species classification problem is based on the combination of two different types of data. Herein, 16S rRNA gene sequence data is used for phylogenetic tree inference and the corresponding binary tree splits are learned based on FAME data. We call this learning approach 'phylogenetic learning'. Supervised Random Forest models are developed to train the classification tasks in a stratified cross-validation setting. In this way, better classification results are obtained for species that are typically hard to distinguish by a single or flat multi-class classification model. Conclusions FAME-based bacterial species classification is successfully evaluated in a taxonomic framework. Although the proposed approach does not improve the overall accuracy compared to flat multi-class classification, it has some distinct advantages. First, it has better capabilities for distinguishing species on which flat multi-class classification fails. Secondly, the hierarchical classification structure allows to easily evaluate and visualize the resolution of FAME data for the discrimination of bacterial species. Summarized, by phylogenetic learning we are able to situate and evaluate FAME-based bacterial species classification in a more informative context. PMID:20113515

Teaching Soil Science in Primary and Secondary Schools

NASA Technical Reports Server (NTRS)

Levine, Elissa R.

1998-01-01

Earth's thin layer of soil is a fragile resource, made up of minerals, organic materials, air, water, and billions of living organisms. Soils plays a variety of critical roles that sustain life on Earth. If we think about soil, we tend to see it first as the source of most of the food we eat and the fibers we use, such as wood and cotton. Few students realize that soils also provide the key ingredients to many of the medicines (including antibiotics), cosmetics, and dyes that we use. Fewer still understand the importance of soils in integrating, controlling, and regulating the movement of air, water, materials, and energy between the hydrosphere, lithosphere, atmosphere, and biosphere. Because soil sustains life, it offers both a context and a natural laboratory for investigating these interactions. The enclosed poster, which integrates soil profiles with typical landscapes in which soils form, can also help students explore the interrelationships of Earth systems and gain an understanding of our soil resources. The poster, produced jointly by the American Geological Institute and the Soil Science Society of America, aims to increase awareness of the importance of soil, as does the GLOBE (Global Learning and Observations To Benefit the Environment) Program. Vice President Al Gore instituted the GLOBE Program on Earth Day of 1993 to increase environmental awareness of individuals throughout the world, contribute to a better scientific understanding of the Earth, and help all students reach higher levels of achievement in science and mathematics. GLOBE functions as a partnership between scientists, students, and teachers in which scientists design protocols for specific measurements they need for their research that can be performed by K-12 students. Teachers are trained in the GLOBE protocols and teach them to their students. Students make the measurements, enter data via the Internet to a central data archive, and the data becomes available to scientists and the general community. Students benefit by having a "hands-on"experience in science, math, and technology, using their local environment as a learning laboratory, as well as contact with scientists and other students around the world. Soil investigations have become an essential component of GLOBE. The protocols that have been developed so far within the GLOBE program include GPS Location, Atmosphere/Climate, Soil Characterization, Soil Moisture and Temperature, Land Cover/Biometry, Hydrology, and Satellite Image Classification. For the GLOBE Soil Characterization Protocol, students explore the physical. chemical, and morphological properties of the soil at their study site. They are asked to dig a pit or use an auger to about 1 meter at at least 2 sites.

[Classification of Priority Area for Soil Environmental Protection Around Water Sources: Method Proposed and Case Demonstration].

PubMed

Li, Lei; Wang, Tie-yu; Wang, Xiaojun; Xiao, Rong-bo; Li, Qi-feng; Peng, Chi; Han, Cun-liang

2016-04-15

Based on comprehensive consideration of soil environmental quality, pollution status of river, environmental vulnerability and the stress of pollution sources, a technical method was established for classification of priority area of soil environmental protection around the river-style water sources. Shunde channel as an important drinking water sources of Foshan City, Guangdong province, was studied as a case, of which the classification evaluation system was set up. In detail, several evaluation factors were selected according to the local conditions of nature, society and economy, including the pollution degree of heavy metals in soil and sediment, soil characteristics, groundwater sensitivity, vegetation coverage, the type and location of pollution sources. Data information was mainly obtained by means of field survey, sampling analysis, and remote sensing interpretation. Afterwards, Analytical Hierarchy Process (AHP) was adopted to decide the weight of each factor. The basic spatial data layers were set up respectively and overlaid based on the weighted summation assessment model in Geographical Information System (GIS), resulting in a classification map of soil environmental protection level in priority area of Shunde channel. Accordingly, the area was classified to three levels named as polluted zone, risky zone and safe zone, which respectively accounted for 6.37%, 60.90% and 32.73% of the whole study area. Polluted zone and risky zone were mainly distributed in Lecong, Longjiang and Leliu towns, with pollutants mainly resulted from the long-term development of aquaculture and the industries containing furniture, plastic constructional materials and textile and clothing. In accordance with the main pollution sources of soil, targeted and differentiated strategies were put forward. The newly established evaluation method could be referenced for the protection and sustainable utilization of soil environment around the water sources.

A surface fuel classification for estimating fire effects

Treesearch

Duncan C. Lutes; Robert E. Keane; John F. Caratti

2009-01-01

We present a classification of duff, litter, fine woody debris, and logs that can be used to stratify a project area into sites with fuel loading that yield significantly different emissions and maximum soil surface temperature. Total particulate matter smaller than 2.5?m in diameter and maximum soil surface temperature were simulated using the First...

Predicting and quantifying soil processes using “geomorphon” landform Classification

USDA-ARS?s Scientific Manuscript database

Soil development and behavior vary spatially at multiple observation scales. Predicting and quantifying soil properties and processes via a catena integrates predictable landscape scale variation relevant to both management decisions and soil survey. Soil maps generally convey variation as a set of ...

Soil texture classification algorithm using RGB characteristics of soil images

USDA-ARS?s Scientific Manuscript database

Soil texture has an important influence on agriculture, affecting crop selection, movement of nutrients and water, soil electrical conductivity, and crop growth. Soil texture has traditionally been determined in the laboratory using pipette and hydrometer methods that require a considerable amount o...

Deep learning for tumor classification in imaging mass spectrometry.

PubMed

Behrmann, Jens; Etmann, Christian; Boskamp, Tobias; Casadonte, Rita; Kriegsmann, Jörg; Maaß, Peter

2018-04-01

Tumor classification using imaging mass spectrometry (IMS) data has a high potential for future applications in pathology. Due to the complexity and size of the data, automated feature extraction and classification steps are required to fully process the data. Since mass spectra exhibit certain structural similarities to image data, deep learning may offer a promising strategy for classification of IMS data as it has been successfully applied to image classification. Methodologically, we propose an adapted architecture based on deep convolutional networks to handle the characteristics of mass spectrometry data, as well as a strategy to interpret the learned model in the spectral domain based on a sensitivity analysis. The proposed methods are evaluated on two algorithmically challenging tumor classification tasks and compared to a baseline approach. Competitiveness of the proposed methods is shown on both tasks by studying the performance via cross-validation. Moreover, the learned models are analyzed by the proposed sensitivity analysis revealing biologically plausible effects as well as confounding factors of the considered tasks. Thus, this study may serve as a starting point for further development of deep learning approaches in IMS classification tasks. https://gitlab.informatik.uni-bremen.de/digipath/Deep_Learning_for_Tumor_Classification_in_IMS. jbehrmann@uni-bremen.de or christianetmann@uni-bremen.de. Supplementary data are available at Bioinformatics online.

Classification/Categorization Model of Instruction for Learning Disabled Students.

ERIC Educational Resources Information Center

Freund, Lisa A.

1987-01-01

Learning-disabled students deficient in classification and categorization require specific instruction in these skills. Use of a classification/categorization instructional model improved the questioning strategies of 60 learning-disabled students, aged 10 to 12. The use of similar models is discussed as a basis for instruction in science, social…

Ch'ol nomenclature for soil classification in the ejido Oxolotán, Tacotalpa, Tabasco, México.

PubMed

Sánchez-Hernández, Rufo; Méndez-De la Cruz, Lucero; Palma-López, David J; Bautista-Zuñiga, Francisco

2018-05-30

The traditional ecological knowledge of land of the Ch'ol originary people from southeast Mexico forms part of their cultural identity; it is local and holistic and implies an integrated physical and spiritual worldview that contributes to improve their living conditions. We analyzed the nomenclature for soil classification used in the Mexican state of Tabasco by the Ch'ol farmers with the objective of contributing to the knowledge of the Maya soil classification. A map of the study area was generated from the digital database of parcels in the ejido Oxolotán in the municipality of Tacotalpa, to which a geopedological map was overlaid in order to obtain modeled topographic profiles (Zavala-Cruz et al., Ecosistemas y Recursos Agropecuarios 3:161-171, 2016). In each modeled profile, a soil profile was made and classified according to IUSS Working Group WRB (181, 2014) in order to generate a map of soil groups, which was used to survey the study area with the participation of 245 local Ch'ol farmers for establishing an ethnopedological soil classification (Ortiz et al.: 62, 1990). In addition, we organized a participatory workshop with 35 people to know details of the names of the soils and their indicators of fertility and workability, from which we selected 15 participants for field trips and description of soil profiles. The color, texture, and stoniness are attributes important in the Ch'ol nomenclature, although the names do not completely reflect the visible characteristic of the soil surface. On the other hand, the mere presence of stones is sufficient to name a land class, while according to IUSS Working Group WRB (181, 2014), a certain amount and distribution of stones in the soil profiles is necessary to be taken into consideration in the name. Perception of soil quality by local farmers considers the compaction or hardness of the cultivable soil layer, because of which black or sandy soils are perceived as better for cultivation of banana, or as secondary vegetation in fallow. Red, yellow, or brown soils are seen as of less quality and are only used for establishing grasslands, while maize is cultivated in all soil classes. Farmers provided the Ch'ol nomenclature, perceived problems, and uses of each class of soil. Translation of Ch'ol soil names and comparison with descriptions of soil profiles revealed that the Ch'ol soil nomenclature takes into account the soil profile, given it is based on characteristics of both surface and subsurface horizons including color of soil matrix and mottles, stoniness, texture, and vegetation.

Active Learning of Classification Models with Likert-Scale Feedback.

PubMed

Xue, Yanbing; Hauskrecht, Milos

2017-01-01

Annotation of classification data by humans can be a time-consuming and tedious process. Finding ways of reducing the annotation effort is critical for building the classification models in practice and for applying them to a variety of classification tasks. In this paper, we develop a new active learning framework that combines two strategies to reduce the annotation effort. First, it relies on label uncertainty information obtained from the human in terms of the Likert-scale feedback. Second, it uses active learning to annotate examples with the greatest expected change. We propose a Bayesian approach to calculate the expectation and an incremental SVM solver to reduce the time complexity of the solvers. We show the combination of our active learning strategy and the Likert-scale feedback can learn classification models more rapidly and with a smaller number of labeled instances than methods that rely on either Likert-scale labels or active learning alone.

Active Learning of Classification Models with Likert-Scale Feedback

PubMed Central

Xue, Yanbing; Hauskrecht, Milos

2017-01-01

Annotation of classification data by humans can be a time-consuming and tedious process. Finding ways of reducing the annotation effort is critical for building the classification models in practice and for applying them to a variety of classification tasks. In this paper, we develop a new active learning framework that combines two strategies to reduce the annotation effort. First, it relies on label uncertainty information obtained from the human in terms of the Likert-scale feedback. Second, it uses active learning to annotate examples with the greatest expected change. We propose a Bayesian approach to calculate the expectation and an incremental SVM solver to reduce the time complexity of the solvers. We show the combination of our active learning strategy and the Likert-scale feedback can learn classification models more rapidly and with a smaller number of labeled instances than methods that rely on either Likert-scale labels or active learning alone. PMID:28979827

A Locality-Constrained and Label Embedding Dictionary Learning Algorithm for Image Classification.

PubMed

Zhengming Li; Zhihui Lai; Yong Xu; Jian Yang; Zhang, David

2017-02-01

Locality and label information of training samples play an important role in image classification. However, previous dictionary learning algorithms do not take the locality and label information of atoms into account together in the learning process, and thus their performance is limited. In this paper, a discriminative dictionary learning algorithm, called the locality-constrained and label embedding dictionary learning (LCLE-DL) algorithm, was proposed for image classification. First, the locality information was preserved using the graph Laplacian matrix of the learned dictionary instead of the conventional one derived from the training samples. Then, the label embedding term was constructed using the label information of atoms instead of the classification error term, which contained discriminating information of the learned dictionary. The optimal coding coefficients derived by the locality-based and label-based reconstruction were effective for image classification. Experimental results demonstrated that the LCLE-DL algorithm can achieve better performance than some state-of-the-art algorithms.

Soil Taxonomy and land evaluation for forest establishment

Treesearch

Haruyoshi Ikawa

1992-01-01

Soil Taxonomy, the United States system of soil classification, can be used for land evaluation for selected purposes. One use is forest establishment in the tropics, and the soil family category is especially functional for this purpose. The soil family is a bionomial name with descriptions usually of soil texture, mineralogy, and soil temperature classes. If the...

Random whole metagenomic sequencing for forensic discrimination of soils.

PubMed

Khodakova, Anastasia S; Smith, Renee J; Burgoyne, Leigh; Abarno, Damien; Linacre, Adrian

2014-01-01

Here we assess the ability of random whole metagenomic sequencing approaches to discriminate between similar soils from two geographically distinct urban sites for application in forensic science. Repeat samples from two parklands in residential areas separated by approximately 3 km were collected and the DNA was extracted. Shotgun, whole genome amplification (WGA) and single arbitrarily primed DNA amplification (AP-PCR) based sequencing techniques were then used to generate soil metagenomic profiles. Full and subsampled metagenomic datasets were then annotated against M5NR/M5RNA (taxonomic classification) and SEED Subsystems (metabolic classification) databases. Further comparative analyses were performed using a number of statistical tools including: hierarchical agglomerative clustering (CLUSTER); similarity profile analysis (SIMPROF); non-metric multidimensional scaling (NMDS); and canonical analysis of principal coordinates (CAP) at all major levels of taxonomic and metabolic classification. Our data showed that shotgun and WGA-based approaches generated highly similar metagenomic profiles for the soil samples such that the soil samples could not be distinguished accurately. An AP-PCR based approach was shown to be successful at obtaining reproducible site-specific metagenomic DNA profiles, which in turn were employed for successful discrimination of visually similar soil samples collected from two different locations.

A study of the utilization of ERTS-1 data from the Wabash River Basin. [soil mapping, crop identification, water resources

NASA Technical Reports Server (NTRS)

Landgrebe, D. A. (Principal Investigator)

1973-01-01

The author has identified the following significant results. In soil association mapping, computerized analysis of ERTS-1 MSS data has yielded images which will prove useful in the ongoing Cooperative Soil Survey program, involving the Soil Conservation Service of USDA and other state and local agencies. In the present mode of operation, a soil survey for a county may take up to 5 years to be completed. Results indicate that a great deal of soils information can be extracted from ERTS-1 data by computer analysis. This information is expected to be very valuable in the premapping conference phase of a soil survey, resulting in more efficient field operations during the actual mapping. In the earth surface features mapping effort it was found that temporal data improved the classification accuracy of forest classification in Tippecanoe County, Indiana. In water resources study a severe scanner look angle effect was observed in the aircraft scanner data of a test lake which was not present in ERTS-1 data of the same site. This effect was greatly accentuated by surface roughness caused by strong winds. Quantitative evaluation of urban features classification in ERTS-1 data was obtained. An 87.1% test accuracy was obtained for eight categories in Marion County, Indiana.

Metric learning for automatic sleep stage classification.

PubMed

Phan, Huy; Do, Quan; Do, The-Luan; Vu, Duc-Lung

2013-01-01

We introduce in this paper a metric learning approach for automatic sleep stage classification based on single-channel EEG data. We show that learning a global metric from training data instead of using the default Euclidean metric, the k-nearest neighbor classification rule outperforms state-of-the-art methods on Sleep-EDF dataset with various classification settings. The overall accuracy for Awake/Sleep and 4-class classification setting are 98.32% and 94.49% respectively. Furthermore, the superior accuracy is achieved by performing classification on a low-dimensional feature space derived from time and frequency domains and without the need for artifact removal as a preprocessing step.

Site Classification using Multichannel Channel Analysis of Surface Wave (MASW) method on Soft and Hard Ground

NASA Astrophysics Data System (ADS)

Ashraf, M. A. M.; Kumar, N. S.; Yusoh, R.; Hazreek, Z. A. M.; Aziman, M.

2018-04-01

Site classification utilizing average shear wave velocity (Vs(30) up to 30 meters depth is a typical parameter. Numerous geophysical methods have been proposed for estimation of shear wave velocity by utilizing assortment of testing configuration, processing method, and inversion algorithm. Multichannel Analysis of Surface Wave (MASW) method is been rehearsed by numerous specialist and professional to geotechnical engineering for local site characterization and classification. This study aims to determine the site classification on soft and hard ground using MASW method. The subsurface classification was made utilizing National Earthquake Hazards Reduction Program (NERHP) and international Building Code (IBC) classification. Two sites are chosen to acquire the shear wave velocity which is in the state of Pulau Pinang for soft soil and Perlis for hard rock. Results recommend that MASW technique can be utilized to spatially calculate the distribution of shear wave velocity (Vs(30)) in soil and rock to characterize areas.

Working with soils: soil science continuing professional development

NASA Astrophysics Data System (ADS)

Hannam, Jacqueline; Thompson, Dick

2017-04-01

The British Society of Soil Science launched the Working with Soils professional competency programme in 2011. This was in response to concerns from practitioners and professionals of a significant skills gap in various sectors that require soil science skills. The programme includes one and two day courses that cover the qualifications, knowledge and skills required of a professional scientist or engineer conducting a range of contract work. All courses qualify for continuing professional development points with various professional practice schemes. Three courses cover the foundations of soil science namely; describing a soil profile, soil classification and understanding soil variability in the field and landscape. Other tailored courses relate to specific skills required from consultants particularly in the planning process where land is assessed for agricultural quality (agricultural land classification). New courses this year include soil handling and restoration that provides practitioners with knowledge of the appropriate management of large volumes of soil that are disturbed during development projects. The courses have so far successfully trained over 100 delegates ranging from PhD students, environmental consultants and government policy advisors.

Vulnerable land ecosystems classification using spatial context and spectral indices

NASA Astrophysics Data System (ADS)

Ibarrola-Ulzurrun, Edurne; Gonzalo-Martín, Consuelo; Marcello, Javier

2017-10-01

Natural habitats are exposed to growing pressure due to intensification of land use and tourism development. Thus, obtaining information on the vegetation is necessary for conservation and management projects. In this context, remote sensing is an important tool for monitoring and managing habitats, being classification a crucial stage. The majority of image classifications techniques are based upon the pixel-based approach. An alternative is the object-based (OBIA) approach, in which a previous segmentation step merges image pixels to create objects that are then classified. Besides, improved results may be gained by incorporating additional spatial information and specific spectral indices into the classification process. The main goal of this work was to implement and assess object-based classification techniques on very-high resolution imagery incorporating spectral indices and contextual spatial information in the classification models. The study area was Teide National Park in Canary Islands (Spain) using Worldview-2 orthoready imagery. In the classification model, two common indices were selected Normalized Difference Vegetation Index (NDVI) and Optimized Soil Adjusted Vegetation Index (OSAVI), as well as two specific Worldview-2 sensor indices, Worldview Vegetation Index and Worldview Soil Index. To include the contextual information, Grey Level Co-occurrence Matrices (GLCM) were used. The classification was performed training a Support Vector Machine with sufficient and representative number of vegetation samples (Spartocytisus supranubius, Pterocephalus lasiospermus, Descurainia bourgaeana and Pinus canariensis) as well as urban, road and bare soil classes. Confusion Matrices were computed to evaluate the results from each classification model obtaining the highest overall accuracy (90.07%) combining both Worldview indices with the GLCM-dissimilarity.

A Biochar Classification System and Associated Test Methods

DOE Office of Scientific and Technical Information (OSTI.GOV)

Camps-Arbestain, Marta; Amonette, James E.; Singh, Balwant

2015-02-18

In this chapter, a biochar classification system related to its use as soil amendment is proposed. This document builds upon previous work and constrains its scope to materials with properties that satisfy the criteria for biochar as defined by either the International Biochar Initiative (IBI) Biochar Standards or the European Biochar Community (EBC) Standards, and it is intended to minimise the need for testing in addition to those required according to the above-mentioned standards. The classification system envisions enabling stakeholders and commercial entities to (i) identify the most suitable biochar to fulfil the requirements for a particular soil and/or land-use,more » and (ii) distinguish the application of biochar for specific niches (e.g., soilless agriculture). It is based on the best current knowledge and the intention is to periodically review and update the document based on new data and knowledge that become available in the scientific literature. The main thrust of this classification system is based on the direct or indirect beneficial effects that biochar provides from its application to soil. We have classified the potential beneficial effects of biochar application to soils into five categories with their corresponding classes, where applicable: (i) carbon (C) storage value, (ii) fertiliser value, (iii) liming value, (iv) particle-size, and (v) use in soil-less agriculture. A summary of recommended test methods is provided at the end of the chapter.« less

Sensitivity analysis of the GEMS soil organic carbon model to land cover land use classification uncertainties under different climate scenarios in Senegal

USGS Publications Warehouse

Dieye, A.M.; Roy, David P.; Hanan, N.P.; Liu, S.; Hansen, M.; Toure, A.

2012-01-01

Spatially explicit land cover land use (LCLU) change information is needed to drive biogeochemical models that simulate soil organic carbon (SOC) dynamics. Such information is increasingly being mapped using remotely sensed satellite data with classification schemes and uncertainties constrained by the sensing system, classification algorithms and land cover schemes. In this study, automated LCLU classification of multi-temporal Landsat satellite data were used to assess the sensitivity of SOC modeled by the Global Ensemble Biogeochemical Modeling System (GEMS). The GEMS was run for an area of 1560 km2 in Senegal under three climate change scenarios with LCLU maps generated using different Landsat classification approaches. This research provides a method to estimate the variability of SOC, specifically the SOC uncertainty due to satellite classification errors, which we show is dependent not only on the LCLU classification errors but also on where the LCLU classes occur relative to the other GEMS model inputs.

Transfer of the nationwide Czech soil survey data to a foreign soil classification - generating input parameters for a process-based soil erosion modelling approach

NASA Astrophysics Data System (ADS)

Beitlerová, Hana; Hieke, Falk; Žížala, Daniel; Kapička, Jiří; Keiser, Andreas; Schmidt, Jürgen; Schindewolf, Marcus

2017-04-01

Process-based erosion modelling is a developing and adequate tool to assess, simulate and understand the complex mechanisms of soil loss due to surface runoff. While the current state of available models includes powerful approaches, a major drawback is given by complex parametrization. A major input parameter for the physically based soil loss and deposition model EROSION 3D is represented by soil texture. However, as the model has been developed in Germany it is dependent on the German soil classification. To exploit data generated during a massive nationwide soil survey campaign taking place in the 1960s across the entire Czech Republic, a transfer from the Czech to the German or at least international (e.g. WRB) system is mandatory. During the survey the internal differentiation of grain sizes was realized in a two fractions approach, separating texture into solely above and below 0.01 mm rather than into clayey, silty and sandy textures. Consequently, the Czech system applies a classification of seven different textures based on the respective percentage of large and small particles, while in Germany 31 groups are essential. The followed approach of matching Czech soil survey data to the German system focusses on semi-logarithmic interpolation of the cumulative soil texture curve additionally on a regression equation based on a recent database of 128 soil pits. Furthermore, for each of the seven Czech texture classes a group of typically suitable classes of the German system was derived. A GIS-based spatial analysis to test approaches of interpolation the soil texture was carried out. First results show promising matches and pave the way to a Czech model application of EROSION 3D.

An Ontology to Support the Classification of Learning Material in an Organizational Learning Environment: An Evaluation

ERIC Educational Resources Information Center

Valaski, Joselaine; Reinehr, Sheila; Malucelli, Andreia

2017-01-01

Purpose: The purpose of this research was to evaluate whether ontology integrated in an organizational learning environment may support the automatic learning material classification in a specific knowledge area. Design/methodology/approach: An ontology for recommending learning material was integrated in the organizational learning environment…

Myths and legends in learning classification rules

NASA Technical Reports Server (NTRS)

Buntine, Wray

1990-01-01

A discussion is presented of machine learning theory on empirically learning classification rules. Six myths are proposed in the machine learning community that address issues of bias, learning as search, computational learning theory, Occam's razor, universal learning algorithms, and interactive learning. Some of the problems raised are also addressed from a Bayesian perspective. Questions are suggested that machine learning researchers should be addressing both theoretically and experimentally.

Applying Active Learning to Assertion Classification of Concepts in Clinical Text

PubMed Central

Chen, Yukun; Mani, Subramani; Xu, Hua

2012-01-01

Supervised machine learning methods for clinical natural language processing (NLP) research require a large number of annotated samples, which are very expensive to build because of the involvement of physicians. Active learning, an approach that actively samples from a large pool, provides an alternative solution. Its major goal in classification is to reduce the annotation effort while maintaining the quality of the predictive model. However, few studies have investigated its uses in clinical NLP. This paper reports an application of active learning to a clinical text classification task: to determine the assertion status of clinical concepts. The annotated corpus for the assertion classification task in the 2010 i2b2/VA Clinical NLP Challenge was used in this study. We implemented several existing and newly developed active learning algorithms and assessed their uses. The outcome is reported in the global ALC score, based on the Area under the average Learning Curve of the AUC (Area Under the Curve) score. Results showed that when the same number of annotated samples was used, active learning strategies could generate better classification models (best ALC – 0.7715) than the passive learning method (random sampling) (ALC – 0.7411). Moreover, to achieve the same classification performance, active learning strategies required fewer samples than the random sampling method. For example, to achieve an AUC of 0.79, the random sampling method used 32 samples, while our best active learning algorithm required only 12 samples, a reduction of 62.5% in manual annotation effort. PMID:22127105

Image Analysis and Classification Based on Soil Strength

DTIC Science & Technology

2016-08-01

Satellite imagery classification is useful for a variety of commonly used ap- plications, such as land use classification, agriculture , wetland...required use of a coinci- dent digital elevation model (DEM) and a high-resolution orthophoto- graph collected by the National Agriculture Imagery Program...14. ABSTRACT Satellite imagery classification is useful for a variety of commonly used applications, such as land use classification, agriculture

MODIS-based spatiotemporal patterns of soil moisture and evapotranspiration interactions in Tampa Bay urban watershed

NASA Astrophysics Data System (ADS)

Chang, Ni-Bin; Xuan, Zhemin; Wimberly, Brent

2011-09-01

Soil moisture and evapotranspiration (ET) is affected by both water and energy balances in the soilvegetation- atmosphere system, it involves many complex processes in the nexus of water and thermal cycles at the surface of the Earth. These impacts may affect the recharge of the upper Floridian aquifer. The advent of urban hydrology and remote sensing technologies opens new and innovative means to undertake eventbased assessment of ecohydrological effects in urban regions. For assessing these landfalls, the multispectral Moderate Resolution Imaging Spectroradiometer (MODIS) remote sensing images can be used for the estimation of such soil moisture change in connection with two other MODIS products - Enhanced Vegetation Index (EVI), Land Surface Temperature (LST). Supervised classification for soil moisture retrieval was performed for Tampa Bay area on the 2 kmx2km grid with MODIS images. Machine learning with genetic programming model for soil moisture estimation shows advances in image processing, feature extraction, and change detection of soil moisture. ET data that were derived by Geostationary Operational Environmental Satellite (GOES) data and hydrologic models can be retrieved from the USGS web site directly. Overall, the derived soil moisture in comparison with ET time series changes on a seasonal basis shows that spatial and temporal variations of soil moisture and ET that are confined within a defined region for each type of surfaces, showing clustered patterns and featuring space scatter plot in association with the land use and cover map. These concomitant soil moisture patterns and ET fluctuations vary among patches, plant species, and, especially, location on the urban gradient. Time series plots of LST in association with ET, soil moisture and EVI reveals unique ecohydrological trends. Such ecohydrological assessment can be applied for supporting the urban landscape management in hurricane-stricken regions.

Demarcation of potentially mineral-deficient areas in central and northern Namibia by means of natural classification systems.

PubMed

Grant, C C; Biggs, H C; Meissner, H H

1996-06-01

Mineral deficiencies that lead to production losses often occur concurrently with climatic and management changes. To diagnose these deficiencies in time to prevent production losses, long-term monitoring of mineral status is advisable. Different classification systems were examined to determine whether areas of possible mineral deficiencies could be identified, so that those which were promising could then be selected for further monitoring purposes. The classification systems addressed differences in soil, vegetation and geology, and were used to define the cattle-ranching areas in the central and northern districts of Namibia. Copper (Cu), Iron (Fe), zinc (Zn), manganese (Mn) and cobalt (Co) concentrations were determined in cattle livers collected at abattoirs. Pooled faecal grab samples and milk samples were collected by farmers, and used to determine phosphorus (P) and calcium (Ca), and iodine (I) status, respectively. Areas of low P concentrations could be identified by all classification systems. The lowest P concentrations were recorded in samples from the Kalahari-sand area, whereas faecal samples collected from cattle on farms in the more arid areas, where the harder soils are mostly found, rarely showed low P concentrations. In the north of the country, low iodine levels were found in milk samples collected from cows grazing on farms in the northern Kalahari broad-leaved woodland. Areas supporting animals with marginal Cu status, could be effectively identified by the detailed soil-classification system of irrigation potential. Copper concentrations were lowest in areas of arid soils, but no indication of Co, Fe, Zn, or Mn deficiencies were found. For most minerals, the geological classification was the best single indicator of areas of lower concentrations. Significant monthly variation for all minerals could also be detected within the classification system. It is concluded that specific classification systems can be useful as indicators of areas with lower mineral concentrations or possible deficiencies.

Deep Learning for ECG Classification

NASA Astrophysics Data System (ADS)

Pyakillya, B.; Kazachenko, N.; Mikhailovsky, N.

2017-10-01

The importance of ECG classification is very high now due to many current medical applications where this problem can be stated. Currently, there are many machine learning (ML) solutions which can be used for analyzing and classifying ECG data. However, the main disadvantages of these ML results is use of heuristic hand-crafted or engineered features with shallow feature learning architectures. The problem relies in the possibility not to find most appropriate features which will give high classification accuracy in this ECG problem. One of the proposing solution is to use deep learning architectures where first layers of convolutional neurons behave as feature extractors and in the end some fully-connected (FCN) layers are used for making final decision about ECG classes. In this work the deep learning architecture with 1D convolutional layers and FCN layers for ECG classification is presented and some classification results are showed.

Comparison of hand-craft feature based SVM and CNN based deep learning framework for automatic polyp classification.

PubMed

Younghak Shin; Balasingham, Ilangko

2017-07-01

Colonoscopy is a standard method for screening polyps by highly trained physicians. Miss-detected polyps in colonoscopy are potential risk factor for colorectal cancer. In this study, we investigate an automatic polyp classification framework. We aim to compare two different approaches named hand-craft feature method and convolutional neural network (CNN) based deep learning method. Combined shape and color features are used for hand craft feature extraction and support vector machine (SVM) method is adopted for classification. For CNN approach, three convolution and pooling based deep learning framework is used for classification purpose. The proposed framework is evaluated using three public polyp databases. From the experimental results, we have shown that the CNN based deep learning framework shows better classification performance than the hand-craft feature based methods. It achieves over 90% of classification accuracy, sensitivity, specificity and precision.

Ethnopedology and soil quality of bamboo (Bambusa sp.) based agroforestry system.

PubMed

Arun Jyoti, Nath; Lal, Rattan; Das, Ashesh Kumar

2015-07-15

It is widely recognized that farmers' hold important knowledge of folk soil classification for agricultural land for its uses, yet little has been studied for traditional agroforestry systems. This article explores the ethnopedology of bamboo (Bambusa sp.) based agroforestry system in North East India, and establishes the relationship of soil quality index (SQI) with bamboo productivity. The study revealed four basic folk soil (mati) types: kalo (black soil), lal (red soil), pathal (stony soil) and balu (sandy soil). Of these, lal mati soil was the most predominant soil type (~ 40%) in bamboo-based agroforestry system. Soil physio-chemical parameters were studied to validate the farmers' soil hierarchal classification and also to correlate with productivity of the bamboo stand. Farmers' hierarchal folk soil classification was consistent with the laboratory scientific analysis. Culm production (i.e. measure of productivity of bamboo) was the highest (27culmsclump(-1)) in kalo mati (black soil) and the lowest (19culmsclump(-1)) in balu mati (sandy soil). Linear correlation of individual soil quality parameter with bamboo productivity explained 16 to 49% of the variability. A multiple correlation of the best fitted linear soil quality parameter (soil organic carbon or SOC, water holding capacity or WHC, total nitrogen) with productivity improved explanatory power to 53%. Development of SQI from ten relevant soil quality parameters and its correlation with bamboo productivity explained the 64% of the variation and therefore, suggest SQI as the best determinant of bamboo yield. Data presented indicate that the kalo mati (black soil) is sustainable or sustainable with high input. However, the other three folk soil types (red, stony and sandy soil) are also sustainable but for other land uses. Therefore, ethnopedological studies may move beyond routine laboratory analysis and incorporate SQI for assessing the sustainability of land uses managed by the farmers'. Additional research is required to incorporate principal component analysis for improving the SQI and site potential assessment. It is also important to evaluate the minimum data set (MDS) required for SQI and productivity assessment in agroforestry systems. Copyright © 2015 Elsevier B.V. All rights reserved.

The generalization ability of online SVM classification based on Markov sampling.

PubMed

Xu, Jie; Yan Tang, Yuan; Zou, Bin; Xu, Zongben; Li, Luoqing; Lu, Yang

2015-03-01

In this paper, we consider online support vector machine (SVM) classification learning algorithms with uniformly ergodic Markov chain (u.e.M.c.) samples. We establish the bound on the misclassification error of an online SVM classification algorithm with u.e.M.c. samples based on reproducing kernel Hilbert spaces and obtain a satisfactory convergence rate. We also introduce a novel online SVM classification algorithm based on Markov sampling, and present the numerical studies on the learning ability of online SVM classification based on Markov sampling for benchmark repository. The numerical studies show that the learning performance of the online SVM classification algorithm based on Markov sampling is better than that of classical online SVM classification based on random sampling as the size of training samples is larger.

Microscale soil structure development after glacial retreat - using machine-learning based segmentation of elemental distributions obtained by NanoSIMS

NASA Astrophysics Data System (ADS)

Schweizer, Steffen; Schlueter, Steffen; Hoeschen, Carmen; Koegel-Knabner, Ingrid; Mueller, Carsten W.

2017-04-01

Soil organic matter (SOM) is distributed on mineral surfaces depending on physicochemical soil properties that vary at the submicron scale. Nanoscale secondary ion mass spectrometry (NanoSIMS) can be used to visualize the spatial distribution of up to seven elements simultaneously at a lateral resolution of approximately 100 nm from which patterns of SOM coatings can be derived. Existing computational methods are mostly confined to visualization and lack spatial quantification measures of coverage and connectivity of organic matter coatings. This study proposes a methodology for the spatial analysis of SOM coatings based on supervised pixel classification and automatic image analysis of the 12C, 12C14N (indicative for SOM) and 16O (indicative for mineral surfaces) secondary ion distributions. The image segmentation of the secondary ion distributions into mineral particle surface and organic coating was done with a machine learning algorithm, which accounts for multiple features like size, color, intensity, edge and texture in all three ion distributions simultaneously. Our workflow allowed the spatial analysis of differences in the SOM coverage during soil development in the Damma glacier forefield (Switzerland) based on NanoSIMS measurements (n=121; containing ca. 4000 particles). The Damma chronosequence comprises several stages of soil development with increasing ice-free period (from ca. 15 to >700 years). To investigate mineral-associated SOM in the developing soil we obtained clay fractions (<2 μm) from two density fractions: light mineral (1.6 to 2.2 g cm3) and heavy mineral (>2.2 g cm3). We found increased coverage and a simultaneous development from patchy-distributed organic coatings to more connected coatings with increasing time after glacial retreat. The normalized N:C ratio (12C14N: (12C14N + 12C)) on the organic matter coatings was higher in the medium-aged soils than in the young and mature ones in both heavy and light mineral fraction. This reflects the sequential accumulation of proteinaceous SOM in the medium-aged soils and C-rich compounds in the mature soils. The results of our microscale image analysis correlated well with the SOM concentration of the fractions measured by elemental analyzer. Image analysis in combination with secondary ion distributions provides a powerful tool at the required microscale and enhances our mechanistic understanding of SOM stabilization in soil.

A study of the utilization of ERTS-1 data from the Wabash River Basin. [crop identification, water resources, urban land use, soil mapping, and atmospheric modeling

NASA Technical Reports Server (NTRS)

Landgrebe, D. A. (Principal Investigator)

1974-01-01

The author has identified the following significant results. The most significant results were obtained in the water resources research, urban land use mapping, and soil association mapping projects. ERTS-1 data was used to classify water bodies to determine acreages and high agreement was obtained with USGS figures. Quantitative evaluation was achieved of urban land use classifications from ERTS-1 data and an overall test accuracy of 90.3% was observed. ERTS-1 data classifications of soil test sites were compared with soil association maps scaled to match the computer produced map and good agreement was observed. In some cases the ERTS-1 results proved to be more accurate than the soil association map.

Myths and legends in learning classification rules

NASA Technical Reports Server (NTRS)

Buntine, Wray

1990-01-01

This paper is a discussion of machine learning theory on empirically learning classification rules. The paper proposes six myths in the machine learning community that address issues of bias, learning as search, computational learning theory, Occam's razor, 'universal' learning algorithms, and interactive learnings. Some of the problems raised are also addressed from a Bayesian perspective. The paper concludes by suggesting questions that machine learning researchers should be addressing both theoretically and experimentally.

Application of GIS-based Procedure on Slopeland Use Classification and Identification

NASA Astrophysics Data System (ADS)

KU, L. C.; LI, M. C.

2016-12-01

In Taiwan, the "Slopeland Conservation and Utilization Act" regulates the management of the slopelands. It categorizes the slopeland into land suitable for agricultural or animal husbandry, land suitable for forestry and land for enhanced conservation, according to the environmental factors of average slope, effective soil depth, soil erosion and parental rock. Traditionally, investigations of environmental factors require cost-effective field works. It has been confronted with many practical issues such as non-evaluated cadastral parcels, evaluation results depending on expert's opinion, difficulties in field measurement and judgment, and time consuming. This study aimed to develop a GIS-based procedure involved in the acceleration of slopeland use classification and quality improvement. First, the environmental factors of slopelands were analyzed by GIS and SPSS software. The analysis involved with the digital elevation model (DEM), soil depth map, land use map and satellite images. Second, 5% of the analyzed slopelands were selected to perform the site investigations and correct the results of classification. Finally, a 2nd examination was involved by randomly selected 2% of the analyzed slopelands to perform the accuracy evaluation. It was showed the developed procedure is effective in slopeland use classification and identification. Keywords: Slopeland Use Classification, GIS, Management

Global Similarity Predicts Dissociation of Classification and Recognition: Evidence Questioning the Implicit-Explicit Learning Distinction in Amnesia

ERIC Educational Resources Information Center

Jamieson, Randall K.; Holmes, Signy; Mewhort, D. J. K.

2010-01-01

Dissociation of classification and recognition in amnesia is widely taken to imply 2 functional systems: an implicit procedural-learning system that is spared in amnesia and an explicit episodic-learning system that is compromised. We argue that both tasks reflect the global similarity of probes to memory. In classification, subjects sort…

The 'Soil Cover App' - a new tool for fast determination of dead and living biomass on soil

NASA Astrophysics Data System (ADS)

Bauer, Thomas; Strauss, Peter; Riegler-Nurscher, Peter; Prankl, Johann; Prankl, Heinrich

2017-04-01

Worldwide many agricultural practices aim on soil protection strategies using living or dead biomass as soil cover. Especially for the case when management practices are focusing on soil erosion mitigation the effectiveness of these practices is directly driven by the amount of soil coverleft on the soil surface. Hence there is a need for quick and reliable methods of soil cover estimation not only for living biomass but particularly for dead biomass (mulch). Available methods for the soil cover measurement are either subjective, depending on an educated guess or time consuming, e.g., if the image is analysed manually at grid points. We therefore developed a mobile application using an algorithm based on entangled forest classification. The final output of the algorithm gives classified labels for each pixel of the input image as well as the percentage of each class which are living biomass, dead biomass, stones and soil. Our training dataset consisted of more than 250 different images and their annotated class information. Images have been taken in a set of different environmental conditions such as light, soil coverages from between 0% to 100%, different materials such as living plants, residues, straw material and stones. We compared the results provided by our mobile application with a data set of 180 images that had been manually annotated A comparison between both methods revealed a regression slope of 0.964 with a coefficient of determination R2 = 0.92, corresponding to an average error of about 4%. While average error of living plant classification was about 3%, dead residue classification resulted in an 8% error. Thus the new mobile application tool offers a fast and easy way to obtain information on the protective potential of a particular agricultural management site.

Perceptual-motor skill learning in Gilles de la Tourette syndrome. Evidence for multiple procedural learning and memory systems.

PubMed

Marsh, Rachel; Alexander, Gerianne M; Packard, Mark G; Zhu, Hongtu; Peterson, Bradley S

2005-01-01

Procedural learning and memory systems likely comprise several skills that are differentially affected by various illnesses of the central nervous system, suggesting their relative functional independence and reliance on differing neural circuits. Gilles de la Tourette syndrome (GTS) is a movement disorder that involves disturbances in the structure and function of the striatum and related circuitry. Recent studies suggest that patients with GTS are impaired in performance of a probabilistic classification task that putatively involves the acquisition of stimulus-response (S-R)-based habits. Assessing the learning of perceptual-motor skills and probabilistic classification in the same samples of GTS and healthy control subjects may help to determine whether these various forms of procedural (habit) learning rely on the same or differing neuroanatomical substrates and whether those substrates are differentially affected in persons with GTS. Therefore, we assessed perceptual-motor skill learning using the pursuit-rotor and mirror tracing tasks in 50 patients with GTS and 55 control subjects who had previously been compared at learning a task of probabilistic classifications. The GTS subjects did not differ from the control subjects in performance of either the pursuit rotor or mirror-tracing tasks, although they were significantly impaired in the acquisition of a probabilistic classification task. In addition, learning on the perceptual-motor tasks was not correlated with habit learning on the classification task in either the GTS or healthy control subjects. These findings suggest that the differing forms of procedural learning are dissociable both functionally and neuroanatomically. The specific deficits in the probabilistic classification form of habit learning in persons with GTS are likely to be a consequence of disturbances in specific corticostriatal circuits, but not the same circuits that subserve the perceptual-motor form of habit learning.

Coastal plain soils and geomorphology: a key to understanding forest hydrology

Treesearch

Thomas M. Williams; Devendra M. Amatya

2016-01-01

In the 1950s, Coile published a simple classification of southeastern coastal soils using three characteristics: drainage class, sub-soil depth, and sub-soil texture. These ideas were used by Warren Stuck and Bill Smith to produce a matrix of soils with drainage class as one ordinate and subsoil texture as the second for the South Carolina coastal plain. Soils...

Particle-size distribution models for the conversion of Chinese data to FAO/USDA system.

PubMed

Shangguan, Wei; Dai, YongJiu; García-Gutiérrez, Carlos; Yuan, Hua

2014-01-01

We investigated eleven particle-size distribution (PSD) models to determine the appropriate models for describing the PSDs of 16349 Chinese soil samples. These data are based on three soil texture classification schemes, including one ISSS (International Society of Soil Science) scheme with four data points and two Katschinski's schemes with five and six data points, respectively. The adjusted coefficient of determination r (2), Akaike's information criterion (AIC), and geometric mean error ratio (GMER) were used to evaluate the model performance. The soil data were converted to the USDA (United States Department of Agriculture) standard using PSD models and the fractal concept. The performance of PSD models was affected by soil texture and classification of fraction schemes. The performance of PSD models also varied with clay content of soils. The Anderson, Fredlund, modified logistic growth, Skaggs, and Weilbull models were the best.

Occurrence of shale soils along the Calabar-Itu highway, Southeastern Nigeria and their implication for the subgrade construction.

PubMed

Ilori, Abidemi Olujide

2016-01-01

This study concerned a stretch of 17 km of a 94-km highway alignment in Southeastern Nigeria that has a high incidence of pavement failure arising from subgrade failure. The subgrade of this section of the roadway is composed of Ekenkpon shale, New Netim marl, and Nkporo shale. Under the Unified Soil Classification System, the shales classify as OH (organic clay) and the marl classifies as MH (inorganic silt). Under the American Association of State and Transportation Officials (AASHTO) M 145 soil classification, all these soils classify as A-7-5 soil. Using the AASHTO M 145 group index, none of these soils was considered suitable as subgrade in its native form. Therefore, cement was investigated as a stabilizing agent. Testing demonstrated that 7, 3 and 12 % by weight were the optimum cement contents to reinforce the Ekenkpon shale, New Netim marl, and Nkporo shale, respectively.

Soils and the soil cover of the Valley of Geysers

NASA Astrophysics Data System (ADS)

Kostyuk, D. N.; Gennadiev, A. N.

2014-06-01

The results of field studies of the soil cover within the tourist part of the Valley of Geysers in Kamchatka performed in 2010 and 2011 are discussed. The morphology of soils, their genesis, and their dependence on the degree of hydrothermal impact are characterized; the soil cover patterns developing in the valley are analyzed. On the basis of the materials provided by the Kronotskii Biospheric Reserve and original field data, the soil map of the valley has been developed. The maps of vegetation conditions, soil temperature at the depth of 15 cm, and slopes of the surface have been used for this purpose together with satellite imagery and field descriptions of reference soil profiles. The legend to the soil map includes nine soil units and seven units of parent materials and their textures. Soil names are given according to the classification developed by I.L. Goldfarb (2005) for the soils of hydrothermal fields. The designation of soil horizons follows the new Classification and Diagnostic System of Russian Soils (2004). It is suggested that a new horizon—a thermometamorphic horizon TRM—can be introduced into this system by analogy with other metamorphic (transformed in situ) horizons distinguished in this system. This horizon is typical of the soils partly or completely transformed by hydrothermal impacts.

Exploring Deep Learning and Transfer Learning for Colonic Polyp Classification

PubMed Central

Uhl, Andreas; Wimmer, Georg; Häfner, Michael

2016-01-01

Recently, Deep Learning, especially through Convolutional Neural Networks (CNNs) has been widely used to enable the extraction of highly representative features. This is done among the network layers by filtering, selecting, and using these features in the last fully connected layers for pattern classification. However, CNN training for automated endoscopic image classification still provides a challenge due to the lack of large and publicly available annotated databases. In this work we explore Deep Learning for the automated classification of colonic polyps using different configurations for training CNNs from scratch (or full training) and distinct architectures of pretrained CNNs tested on 8-HD-endoscopic image databases acquired using different modalities. We compare our results with some commonly used features for colonic polyp classification and the good results suggest that features learned by CNNs trained from scratch and the “off-the-shelf” CNNs features can be highly relevant for automated classification of colonic polyps. Moreover, we also show that the combination of classical features and “off-the-shelf” CNNs features can be a good approach to further improve the results. PMID:27847543

Distance Metric Learning via Iterated Support Vector Machines.

PubMed

Zuo, Wangmeng; Wang, Faqiang; Zhang, David; Lin, Liang; Huang, Yuchi; Meng, Deyu; Zhang, Lei

2017-07-11

Distance metric learning aims to learn from the given training data a valid distance metric, with which the similarity between data samples can be more effectively evaluated for classification. Metric learning is often formulated as a convex or nonconvex optimization problem, while most existing methods are based on customized optimizers and become inefficient for large scale problems. In this paper, we formulate metric learning as a kernel classification problem with the positive semi-definite constraint, and solve it by iterated training of support vector machines (SVMs). The new formulation is easy to implement and efficient in training with the off-the-shelf SVM solvers. Two novel metric learning models, namely Positive-semidefinite Constrained Metric Learning (PCML) and Nonnegative-coefficient Constrained Metric Learning (NCML), are developed. Both PCML and NCML can guarantee the global optimality of their solutions. Experiments are conducted on general classification, face verification and person re-identification to evaluate our methods. Compared with the state-of-the-art approaches, our methods can achieve comparable classification accuracy and are efficient in training.

Bruno Braunerde und die Bodentypen - Learning about soil diversity and soil functions with cartoon characters

NASA Astrophysics Data System (ADS)

Hofmann, Anett

2015-04-01

"Bruno Braunerde und die Bodentypen" is a German-language learning material that fosters discovery of soil diversity and soil functions in kids, teens and adults who enjoy interactive learning activities. The learning material consists of (i) a large poster (dimensions 200 x 120 cm) showing an imaginative illustrated landscape that could be situated in Austria, Switzerland or southern Germany and (ii) a set of 15 magnetic cards that show different soil cartoon characters, e.g. Bruno Braunerde (Cambisol), Stauni Pseudogley (Stagnic Luvisol) or Heidi Podsol (Podzol) on the front and a fun profession and address (linked to the respective soil functions) on the back side. The task is to place the soil cartoon characters to their 'home' in the landscape. This learning material was developed as a contribution to the International Year of Soils 2015 and is supported by the German, Austrian and Swiss Soil Sciences Societies and the Swiss Federal Office for the Environment. The soil cartoon characters are an adaptation of the original concept by the James Hutton Institute, Aberdeen, Scotland (www.hutton.ac.uk/learning/dirt-doctor).

Intrinsic Remediation Engineering Evaluation/Cost Analysis for Car Care Center at Bolling Air Force Base, Washington, DC

DTIC Science & Technology

1997-01-01

supplemented using established literature values for similar aquifer materials . The groundwater sampling activities and analytical results from both...subsurface materials recovered. Observed soil classification types compared very favorably to the soil classifications determined by the CPT tests. 0 2.1.5...other similar substances were handled in a manner consistent with accepted safety procedures and standard operating practices. Well completion materials

Impacts of soil moisture content on visual soil evaluation

NASA Astrophysics Data System (ADS)

Emmet-Booth, Jeremy; Forristal, Dermot; Fenton, Owen; Bondi, Giulia; Creamer, Rachel; Holden, Nick

2017-04-01

Visual Soil Examination and Evaluation (VSE) techniques offer tools for soil quality assessment. They involve the visual and tactile assessment of soil properties such as aggregate size and shape, porosity, redox morphology, soil colour and smell. An increasing body of research has demonstrated the reliability and utility of VSE techniques. However a number of limitations have been identified, including the potential impact of soil moisture variation during sampling. As part of a national survey of grassland soil quality in Ireland, an evaluation of the impact of soil moisture on two widely used VSE techniques was conducted. The techniques were Visual Evaluation of Soil Structure (VESS) (Guimarães et al., 2011) and Visual Soil Assessment (VSA) (Shepherd, 2009). Both generate summarising numeric scores that indicate soil structural quality, though employ different scoring mechanisms. The former requires the assessment of properties concurrently and the latter separately. Both methods were deployed on 20 sites across Ireland representing a range of soils. Additional samples were taken for soil volumetric water (θ) determination at 5-10 and 10-20 cm depth. No significant correlation was observed between θ 5-10 cm and either VSE technique. However, VESS scores were significantly related to θ 10-20 cm (rs = 0.40, sig = 0.02) while VSA scores were not (rs = -0.33, sig = 0.06). VESS and VSA scores can be grouped into quality classifications (good, moderate and poor). No significant mean difference was observed between θ 5-10 cm or θ 10-20 cm according to quality classification by either method. It was concluded that VESS scores may be affected by soil moisture variation while VSA appear unaffected. The different scoring mechanisms, where the separate assessment and scoring of individual properties employed by VSA, may limit soil moisture effects. However, moisture content appears not to affect overall structural quality classification by either method. References Guimarães, R.M.C., Ball, B.C. & Tormena, C.A. 2011. Improvements in the visual evaluation of soil structure, Soil Use and Management, 27, 3: 395-403 Shepherd, G.T. 2009. Visual Soil Assessment. Field guide for pastoral grazing and cropping on flat to rolling country. 2nd edn. Horizons regional council, New Zealand.

Effect of e-learning program on risk assessment and pressure ulcer classification - A randomized study.

PubMed

Bredesen, Ida Marie; Bjøro, Karen; Gunningberg, Lena; Hofoss, Dag

2016-05-01

Pressure ulcers (PUs) are a problem in health care. Staff competency is paramount to PU prevention. Education is essential to increase skills in pressure ulcer classification and risk assessment. Currently, no pressure ulcer learning programs are available in Norwegian. Develop and test an e-learning program for assessment of pressure ulcer risk and pressure ulcer classification. Forty-four nurses working in acute care hospital wards or nursing homes participated and were assigned randomly into two groups: an e-learning program group (intervention) and a traditional classroom lecture group (control). Data was collected immediately before and after training, and again after three months. The study was conducted at one nursing home and two hospitals between May and December 2012. Accuracy of risk assessment (five patient cases) and pressure ulcer classification (40 photos [normal skin, pressure ulcer categories I-IV] split in two sets) were measured by comparing nurse evaluations in each of the two groups to a pre-established standard based on ratings by experts in pressure ulcer classification and risk assessment. Inter-rater reliability was measured by exact percent agreement and multi-rater Fleiss kappa. A Mann-Whitney U test was used for continuous sum score variables. An e-learning program did not improve Braden subscale scoring. For pressure ulcer classification, however, the intervention group scored significantly higher than the control group on several of the categories in post-test immediately after training. However, after three months there were no significant differences in classification skills between the groups. An e-learning program appears to have a greater effect on the accuracy of pressure ulcer classification than classroom teaching in the short term. For proficiency in Braden scoring, no significant effect of educational methods on learning results was detected. Copyright © 2016 Elsevier Ltd. All rights reserved.

Learning classification with auxiliary probabilistic information

PubMed Central

Nguyen, Quang; Valizadegan, Hamed; Hauskrecht, Milos

2012-01-01

Finding ways of incorporating auxiliary information or auxiliary data into the learning process has been the topic of active data mining and machine learning research in recent years. In this work we study and develop a new framework for classification learning problem in which, in addition to class labels, the learner is provided with an auxiliary (probabilistic) information that reflects how strong the expert feels about the class label. This approach can be extremely useful for many practical classification tasks that rely on subjective label assessment and where the cost of acquiring additional auxiliary information is negligible when compared to the cost of the example analysis and labelling. We develop classification algorithms capable of using the auxiliary information to make the learning process more efficient in terms of the sample complexity. We demonstrate the benefit of the approach on a number of synthetic and real world data sets by comparing it to the learning with class labels only. PMID:25309141

Digital soil classification and elemental mapping using imaging Vis-NIR spectroscopy: How to explicitly quantify stagnic properties of a Luvisol under Norway spruce

NASA Astrophysics Data System (ADS)

Kriegs, Stefanie; Buddenbaum, Henning; Rogge, Derek; Steffens, Markus

2015-04-01

Laboratory imaging Vis-NIR spectroscopy of soil profiles is a novel technique in soil science that can determine quantity and quality of various chemical soil properties with a hitherto unreached spatial resolution in undisturbed soil profiles. We have applied this technique to soil cores in order to get quantitative proof of redoximorphic processes under two different tree species and to proof tree-soil interactions at microscale. Due to the imaging capabilities of Vis-NIR spectroscopy a spatially explicit understanding of soil processes and properties can be achieved. Spatial heterogeneity of the soil profile can be taken into account. We took six 30 cm long rectangular soil columns of adjacent Luvisols derived from quaternary aeolian sediments (Loess) in a forest soil near Freising/Bavaria using stainless steel boxes (100×100×300 mm). Three profiles were sampled under Norway spruce and three under European beech. A hyperspectral camera (VNIR, 400-1000 nm in 160 spectral bands) with spatial resolution of 63×63 µm² per pixel was used for data acquisition. Reference samples were taken at representative spots and analysed for organic carbon (OC) quantity and quality with a CN elemental analyser and for iron oxides (Fe) content using dithionite extraction followed by ICP-OES measurement. We compared two supervised classification algorithms, Spectral Angle Mapper and Maximum Likelihood, using different sets of training areas and spectral libraries. As established in chemometrics we used multivariate analysis such as partial least-squares regression (PLSR) in addition to multivariate adaptive regression splines (MARS) to correlate chemical data with Vis-NIR spectra. As a result elemental mapping of Fe and OC within the soil core at high spatial resolution has been achieved. The regression model was validated by a new set of reference samples for chemical analysis. Digital soil classification easily visualizes soil properties within the soil profiles. By combining both techniques, detailed soil maps, elemental balances and a deeper understanding of soil forming processes at the microscale become feasible for complete soil profiles.

Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology.

PubMed

Sharma, Harshita; Zerbe, Norman; Klempert, Iris; Hellwich, Olaf; Hufnagl, Peter

2017-11-01

Deep learning using convolutional neural networks is an actively emerging field in histological image analysis. This study explores deep learning methods for computer-aided classification in H&E stained histopathological whole slide images of gastric carcinoma. An introductory convolutional neural network architecture is proposed for two computerized applications, namely, cancer classification based on immunohistochemical response and necrosis detection based on the existence of tumor necrosis in the tissue. Classification performance of the developed deep learning approach is quantitatively compared with traditional image analysis methods in digital histopathology requiring prior computation of handcrafted features, such as statistical measures using gray level co-occurrence matrix, Gabor filter-bank responses, LBP histograms, gray histograms, HSV histograms and RGB histograms, followed by random forest machine learning. Additionally, the widely known AlexNet deep convolutional framework is comparatively analyzed for the corresponding classification problems. The proposed convolutional neural network architecture reports favorable results, with an overall classification accuracy of 0.6990 for cancer classification and 0.8144 for necrosis detection. Copyright © 2017 Elsevier Ltd. All rights reserved.

Classification Framework for ICT-Based Learning Technologies for Disabled People

ERIC Educational Resources Information Center

Hersh, Marion

2017-01-01

The paper presents the first systematic approach to the classification of inclusive information and communication technologies (ICT)-based learning technologies and ICT-based learning technologies for disabled people which covers both assistive and general learning technologies, is valid for all disabled people and considers the full range of…

Online Learning for Classification of Alzheimer Disease based on Cortical Thickness and Hippocampal Shape Analysis.

PubMed

Lee, Ga-Young; Kim, Jeonghun; Kim, Ju Han; Kim, Kiwoong; Seong, Joon-Kyung

2014-01-01

Mobile healthcare applications are becoming a growing trend. Also, the prevalence of dementia in modern society is showing a steady growing trend. Among degenerative brain diseases that cause dementia, Alzheimer disease (AD) is the most common. The purpose of this study was to identify AD patients using magnetic resonance imaging in the mobile environment. We propose an incremental classification for mobile healthcare systems. Our classification method is based on incremental learning for AD diagnosis and AD prediction using the cortical thickness data and hippocampus shape. We constructed a classifier based on principal component analysis and linear discriminant analysis. We performed initial learning and mobile subject classification. Initial learning is the group learning part in our server. Our smartphone agent implements the mobile classification and shows various results. With use of cortical thickness data analysis alone, the discrimination accuracy was 87.33% (sensitivity 96.49% and specificity 64.33%). When cortical thickness data and hippocampal shape were analyzed together, the achieved accuracy was 87.52% (sensitivity 96.79% and specificity 63.24%). In this paper, we presented a classification method based on online learning for AD diagnosis by employing both cortical thickness data and hippocampal shape analysis data. Our method was implemented on smartphone devices and discriminated AD patients for normal group.

Development of a channel classification to evaluate potential for cottonwood restoration, lower segments of the Middle Missouri River, South Dakota and Nebraska

USGS Publications Warehouse

Jacobson, Robert B.; Elliott, Caroline M.; Huhmann, Brittany L.

2010-01-01

This report documents development of a spatially explicit river and flood-plain classification to evaluate potential for cottonwood restoration along the Sharpe and Fort Randall segments of the Middle Missouri River. This project involved evaluating existing topographic, water-surface elevation, and soils data to determine if they were sufficient to create a classification similar to the Land Capability Potential Index (LCPI) developed by Jacobson and others (U.S. Geological Survey Scientific Investigations Report 2007–5256) and developing a geomorphically based classification to apply to evaluating restoration potential.Existing topographic, water-surface elevation, and soils data for the Middle Missouri River were not sufficient to replicate the LCPI. The 1/3-arc-second National Elevation Dataset delineated most of the topographic complexity and produced cumulative frequency distributions similar to a high-resolution 5-meter topographic dataset developed for the Lower Missouri River. However, lack of bathymetry in the National Elevation Dataset produces a potentially critical bias in evaluation of frequently flooded surfaces close to the river. High-resolution soils data alone were insufficient to replace the information content of the LCPI. In test reaches in the Lower Missouri River, soil drainage classes from the Soil Survey Geographic Database database correctly classified 0.8–98.9 percent of the flood-plain area at or below the 5-year return interval flood stage depending on state of channel incision; on average for river miles 423–811, soil drainage class correctly classified only 30.2 percent of the flood-plain area at or below the 5-year return interval flood stage. Lack of congruence between soil characteristics and present-day hydrology results from relatively rapid incision and aggradation of segments of the Missouri River resulting from impoundments and engineering. The most sparsely available data in the Middle Missouri River were water-surface elevations. Whereas hydraulically modeled water-surface elevations were available at 1.6-kilometer intervals in the Lower Missouri River, water-surface elevations in the Middle Missouri River had to be interpolated between streamflow-gaging stations spaced 3–116 kilometers. Lack of high-resolution water-surface elevation data precludes development of LCPI-like classification maps.An hierarchical river classification framework is proposed to provide structure for a multiscale river classification. The segment-scale classification presented in this report is deductive and based on presumed effects of dams, significant tributaries, and geological (and engineered) channel constraints. An inductive reach-scale classification, nested within the segment scale, is based on multivariate statistical clustering of geomorphic data collected at 500-meter intervals along the river. Cluster-based classifications delineate reaches of the river with similar channel and flood-plain geomorphology, and presumably, similar geomorphic and hydrologic processes. The dominant variables in the clustering process were channel width (Fort Randall) and valley width (Sharpe), followed by braiding index (both segments).Clusters with multithread and highly sinuous channels are likely to be associated with dynamic channel migration and deposition of fresh, bare sediment conducive to natural cottonwood germination. However, restoration potential within these reaches is likely to be mitigated by interaction of cottonwood life stages with the highly altered flow regime.

Joint Concept Correlation and Feature-Concept Relevance Learning for Multilabel Classification.

PubMed

Zhao, Xiaowei; Ma, Zhigang; Li, Zhi; Li, Zhihui

2018-02-01

In recent years, multilabel classification has attracted significant attention in multimedia annotation. However, most of the multilabel classification methods focus only on the inherent correlations existing among multiple labels and concepts and ignore the relevance between features and the target concepts. To obtain more robust multilabel classification results, we propose a new multilabel classification method aiming to capture the correlations among multiple concepts by leveraging hypergraph that is proved to be beneficial for relational learning. Moreover, we consider mining feature-concept relevance, which is often overlooked by many multilabel learning algorithms. To better show the feature-concept relevance, we impose a sparsity constraint on the proposed method. We compare the proposed method with several other multilabel classification methods and evaluate the classification performance by mean average precision on several data sets. The experimental results show that the proposed method outperforms the state-of-the-art methods.

Quantum Ensemble Classification: A Sampling-Based Learning Control Approach.

PubMed

Chen, Chunlin; Dong, Daoyi; Qi, Bo; Petersen, Ian R; Rabitz, Herschel

2017-06-01

Quantum ensemble classification (QEC) has significant applications in discrimination of atoms (or molecules), separation of isotopes, and quantum information extraction. However, quantum mechanics forbids deterministic discrimination among nonorthogonal states. The classification of inhomogeneous quantum ensembles is very challenging, since there exist variations in the parameters characterizing the members within different classes. In this paper, we recast QEC as a supervised quantum learning problem. A systematic classification methodology is presented by using a sampling-based learning control (SLC) approach for quantum discrimination. The classification task is accomplished via simultaneously steering members belonging to different classes to their corresponding target states (e.g., mutually orthogonal states). First, a new discrimination method is proposed for two similar quantum systems. Then, an SLC method is presented for QEC. Numerical results demonstrate the effectiveness of the proposed approach for the binary classification of two-level quantum ensembles and the multiclass classification of multilevel quantum ensembles.

Digital soil map of the Ussuri River basin

NASA Astrophysics Data System (ADS)

Bugaets, A. N.; Pschenichnikova, N. F.; Tereshkina, A. A.; Krasnopeev, S. M.; Gartsman, B. I.; Golodnaya, O. M.; Oznobikhin, V. I.

2017-08-01

On the basis of digital soil, topographic, and geological maps; raster topography model; forestry materials; and literature data, the digital soil map of the Ussuri River basin (24400 km2) was created on a scale of 1: 100000. To digitize the initial paper-based maps and analyze the results, an ESRI ArcGIS Desktop (ArcEditor) v.10.1 (http://www.esri.com) and an open-code SAGA GIS v.2.3 (System for Automated Geoscientific Analyses, http://www.saga-gis.org) were used. The spatial distribution of soil areas on the obtained digital soil map is in agreement with modern cartographic data and the SRTM digital elevation model (SRTM DEM). The regional soil classification developed by G.I. Ivanov was used in the legend to the soil map. The names of soil units were also correlated with the names suggested in the modern Russian soil classification system. The major soil units on the map are at the soil subtypes that reflect the entire vertical spectrum of soils in the south of the Far East of Russia (Primorye region). These are mountainous tundra soils, podzolic soils, brown taiga soils, mountainous brown forest soils, bleached brown soils, meadow-brown soils, meadow gley soils, and floodplain soils). With the help of the spatial analysis function of GIS, the comparison of the particular characteristics of the soil cover with numerical characteristics of the topography, geological composition of catchments, and vegetation cover was performed.

Geological Engineering Characteristics of the Residual Soil: Implementation for Soil Bearing Capacity at Gayungan, Surabaya, East Java

NASA Astrophysics Data System (ADS)

Rukmana, Y. Y.; Ridwan, M.

2018-01-01

This paper presents the results of soil investigation on the residual soil at Gayungan Surabaya. The methodology of the research consists of Drilling + Standard Penetration Test (ASTM D1586-99), sampling and laboratory test for index properties & mechanical of soil, then analyzed for Soil Bearing Capacity (Meyerhoff, 1976). Field test analysis data showed that Bore Hole.01(BH.01) and Bore Hole.03 (BH.03) were dominated by Sand / Sandy clay layer with Standart Penetration Test (SPT) values: 6-68, whereas in BH.02 was dominated by Clayey sand layer with Standard Penetration Test (SPT) values: 32-68. Based on Soil classification according to Unified Soil Classification System (USCS), the soil type at the research area consisted of ML (Silt with Low plasticity), CL ( Clay with low plasticity), MH (Silt with High plasticity), and SP (Sand with Poor gradation). Based on the borlog data and soil bearing capacity analysis of the research area is recommended: for The Deep foundation to reaches at least 16 meters depth with Qa = 1160.40-2032.80 kN / m2, and Shallow foundation reaches at least 1-2 meters deep with Qa = 718.25 kN / M2.

Virtual Soil Monoliths: Blending Traditional and Web-Based Educational Approaches

ERIC Educational Resources Information Center

Krzic, Maja; Strivelli, Rachel A.; Holmes, Emma; Grand, Stephanie; Dyanatkar, Saeed; Lavkulich, Les M.; Crowley, Chris

2013-01-01

Since soil plays a crucial role in all aspects of global environmental change, it is essential that post-secondary institutions provide students with a strong foundation in soil science concepts including soil classification. The onset of information technology (IT) and web-based multimedia have opened new avenues to better incorporate…

Abnormality detection of mammograms by discriminative dictionary learning on DSIFT descriptors.

PubMed

Tavakoli, Nasrin; Karimi, Maryam; Nejati, Mansour; Karimi, Nader; Reza Soroushmehr, S M; Samavi, Shadrokh; Najarian, Kayvan

2017-07-01

Detection and classification of breast lesions using mammographic images are one of the most difficult studies in medical image processing. A number of learning and non-learning methods have been proposed for detecting and classifying these lesions. However, the accuracy of the detection/classification still needs improvement. In this paper we propose a powerful classification method based on sparse learning to diagnose breast cancer in mammograms. For this purpose, a supervised discriminative dictionary learning approach is applied on dense scale invariant feature transform (DSIFT) features. A linear classifier is also simultaneously learned with the dictionary which can effectively classify the sparse representations. Our experimental results show the superior performance of our method compared to existing approaches.

Trophic position of soil nematodes in boreal forests as indicated by stable isotope analysis

NASA Astrophysics Data System (ADS)

Kudrin, Alexey; Tsurikov, Sergey

2016-04-01

Despite the well-developed trophic classification of soil nematodes, their position in soil food webs is still little understood. Observed deviations from the typical feeding strategy indicate that a simplified trophic classification probably does not fully reflect actual trophic interactions. Furthermore, the extent and functional significance of nematodes as prey for other soil animals remains unknown. Stable isotope analysis (SIA) is powerful tool for investigating the structure of soil food webs, but its application to the study of soil nematodes has been limited to only a few studies. We used stable isotope analysis to gain a better understanding of trophic links of several groups of soil nematodes in two boreal forests on albeluvisol. We investigated four taxonomic groups of nematodes: Mononchida, Dorylaimida, Plectidae and Tylenchidae (mostly from the genus Filenchus), that according to the conventional trophic classification represent predators, omnivores, bacterivores and root-fungal feeders, respectively. To assess the trophic position of nematodes, we used a comparison against a set of reference species including herbivorous, saprophagous and predatory macro-invertebrates, oribatid and mesostigmatid mites, and collembolans. Our results suggest that trophic position of the investigated groups of soil nematodes generally corresponds to the conventional classification. All nematodes were enriched in 13C relative to Picea abies roots and litter, and mycorrhizal fungal mycelium. Root-fungal feeders Tylenchidae had δ15N values similar to those of earthworms, enchytraeids and Entomobrya collembolans, but slightly lower δ13C values. Bacterivorous Plectidae were either equal or enriched in 15N compared with saprophagous macroinvertebrates and most mesofauna species. Omnivorous Dorylaimida and predatory Mononchida were further enriched in 15N and their isotopic signature was similar to that of predatory arthropods. These data confirm a clear separation of nematodes into saprophagous/microbial feeders (Tylenchidae and Plectidae) and predators (Mononchida and Dorylaimida). Furthermore, they suggest that Mononchida and Dorylaimida use different sources of carbon, though exact trophic links remain unclear. As a rule, nematodes were either equal or higher in δ15N values relative to most microbivorous microarthropods, contradicting an emerging view that soil nematodes can be an important prey for a wide range of oribatid mites and collembolans. Patterns of the isotopic signatures suggest that soil nematodes and the bulk of soil animals depend on resources derived from a dominating upper-canopy tree (Picea abies) via the detrital, rather than mycorrhizal pathway.

Text Classification for Intelligent Portfolio Management

DTIC Science & Technology

2002-05-01

years including nearest neighbor classification [15], naive Bayes with EM (Ex- pectation Maximization) [11] [13], Winnow with active learning [10... Active Learning and Expectation Maximization (EM). In particular, active learning is used to actively select documents for labeling, then EM assigns...generalization with active learning . Machine Learning, 15(2):201–221, 1994. [3] I. Dagan and P. Engelson. Committee-based sampling for training

Do pre-trained deep learning models improve computer-aided classification of digital mammograms?

NASA Astrophysics Data System (ADS)

Aboutalib, Sarah S.; Mohamed, Aly A.; Zuley, Margarita L.; Berg, Wendie A.; Luo, Yahong; Wu, Shandong

2018-02-01

Digital mammography screening is an important exam for the early detection of breast cancer and reduction in mortality. False positives leading to high recall rates, however, results in unnecessary negative consequences to patients and health care systems. In order to better aid radiologists, computer-aided tools can be utilized to improve distinction between image classifications and thus potentially reduce false recalls. The emergence of deep learning has shown promising results in the area of biomedical imaging data analysis. This study aimed to investigate deep learning and transfer learning methods that can improve digital mammography classification performance. In particular, we evaluated the effect of pre-training deep learning models with other imaging datasets in order to boost classification performance on a digital mammography dataset. Two types of datasets were used for pre-training: (1) a digitized film mammography dataset, and (2) a very large non-medical imaging dataset. By using either of these datasets to pre-train the network initially, and then fine-tuning with the digital mammography dataset, we found an increase in overall classification performance in comparison to a model without pre-training, with the very large non-medical dataset performing the best in improving the classification accuracy.

Using greenhouse gas fluxes to define soil functional types

DOE Office of Scientific and Technical Information (OSTI.GOV)

Petrakis, Sandra; Barba, Josep; Bond-Lamberty, Ben

Soils provide key ecosystem services and directly control ecosystem functions; thus, there is a need to define the reference state of soil functionality. Most common functional classifications of ecosystems are vegetation-centered and neglect soil characteristics and processes. We propose Soil Functional Types (SFTs) as a conceptual approach to represent and describe the functionality of soils based on characteristics of their greenhouse gas (GHG) flux dynamics. We used automated measurements of CO2, CH4 and N2O in a forested area to define SFTs following a simple statistical framework. This study supports the hypothesis that SFTs provide additional insights on the spatial variabilitymore » of soil functionality beyond information represented by commonly measured soil parameters (e.g., soil moisture, soil temperature, litter biomass). We discuss the implications of this framework at the plot-scale and the potential of this approach at larger scales. This approach is a first step to provide a framework to define SFTs, but a community effort is necessary to harmonize any global classification for soil functionality. A global application of the proposed SFT framework will only be possible if there is a community-wide effort to share data and create a global database of GHG emissions from soils.« less

SOIL Geo-Wiki: A tool for improving soil information

NASA Astrophysics Data System (ADS)

Skalský, Rastislav; Balkovic, Juraj; Fritz, Steffen; See, Linda; van der Velde, Marijn; Obersteiner, Michael

2014-05-01

Crowdsourcing is increasingly being used as a way of collecting data for scientific research, e.g. species identification, classification of galaxies and unravelling of protein structures. The WorldSoilProfiles.org database at ISRIC is a global collection of soil profiles, which have been 'crowdsourced' from experts. This system, however, requires contributors to have a priori knowledge about soils. Yet many soil parameters can be observed in the field without specific knowledge or equipment such as stone content, soil depth or color. By crowdsourcing this information over thousands of locations, the uncertainty in current soil datasets could be radically reduced, particularly in areas currently without information or where multiple interpretations are possible from different existing soil maps. Improved information on soils could benefit many research fields and applications. Better soil data could enhance assessments of soil ecosystem services (e.g. soil carbon storage) and facilitate improved process-based ecosystem modeling from local to global scales. Geo-Wiki is a crowdsourcing tool that was developed at IIASA for land cover validation using satellite imagery. Several branches are now available focused on specific aspects of land cover validation, e.g. validating cropland extent or urbanized areas. Geo-Wiki Pictures is a smart phone application for collecting land cover related information on the ground. The extension of Geo-Wiki to a mobile environment provides a tool for experts in land cover validation but is also a way of reaching the general public in the validation of land cover. Here we propose a Soil Geo-Wiki tool that builds on the existing functionality of the Geo-Wiki application, which will be largely designed for the collection and sharing of soil information. Two distinct applications are envisaged: an expert-oriented application mainly for scientific purposes, which will use soil science related language (e.g. WRB or any other global reference soil classification system) and allow experts to upload and share scientifically rigorous soil data; and an application oriented towards the general public, which will be more focused on describing well observed, individual soil properties using simplified classification keys. The latter application will avoid the use of soil science related terminology and focus on the most useful soil parameters such as soil surface features, stone content, soil texture, soil plasticity, calcium carbonate presence, soil color, soil pH, soil repellency, and soil depth. Collection of soil and landscape pictures will also be supported in Soil Geo-Wiki to allow for comprehensive data collection while simultaneously allowing for quality checking by experts.

A supervised learning rule for classification of spatiotemporal spike patterns.

PubMed

Lilin Guo; Zhenzhong Wang; Adjouadi, Malek

2016-08-01

This study introduces a novel supervised algorithm for spiking neurons that take into consideration synapse delays and axonal delays associated with weights. It can be utilized for both classification and association and uses several biologically influenced properties, such as axonal and synaptic delays. This algorithm also takes into consideration spike-timing-dependent plasticity as in Remote Supervised Method (ReSuMe). This paper focuses on the classification aspect alone. Spiked neurons trained according to this proposed learning rule are capable of classifying different categories by the associated sequences of precisely timed spikes. Simulation results have shown that the proposed learning method greatly improves classification accuracy when compared to the Spike Pattern Association Neuron (SPAN) and the Tempotron learning rule.

Wishart Deep Stacking Network for Fast POLSAR Image Classification.

PubMed

Jiao, Licheng; Liu, Fang

2016-05-11

Inspired by the popular deep learning architecture - Deep Stacking Network (DSN), a specific deep model for polarimetric synthetic aperture radar (POLSAR) image classification is proposed in this paper, which is named as Wishart Deep Stacking Network (W-DSN). First of all, a fast implementation of Wishart distance is achieved by a special linear transformation, which speeds up the classification of POLSAR image and makes it possible to use this polarimetric information in the following Neural Network (NN). Then a single-hidden-layer neural network based on the fast Wishart distance is defined for POLSAR image classification, which is named as Wishart Network (WN) and improves the classification accuracy. Finally, a multi-layer neural network is formed by stacking WNs, which is in fact the proposed deep learning architecture W-DSN for POLSAR image classification and improves the classification accuracy further. In addition, the structure of WN can be expanded in a straightforward way by adding hidden units if necessary, as well as the structure of the W-DSN. As a preliminary exploration on formulating specific deep learning architecture for POLSAR image classification, the proposed methods may establish a simple but clever connection between POLSAR image interpretation and deep learning. The experiment results tested on real POLSAR image show that the fast implementation of Wishart distance is very efficient (a POLSAR image with 768000 pixels can be classified in 0.53s), and both the single-hidden-layer architecture WN and the deep learning architecture W-DSN for POLSAR image classification perform well and work efficiently.

Automatic Classification Using Supervised Learning in a Medical Document Filtering Application.

ERIC Educational Resources Information Center

Mostafa, J.; Lam, W.

2000-01-01

Presents a multilevel model of the information filtering process that permits document classification. Evaluates a document classification approach based on a supervised learning algorithm, measures the accuracy of the algorithm in a neural network that was trained to classify medical documents on cell biology, and discusses filtering…

Site effect classification based on microtremor data analysis using a concentration-area fractal model

NASA Astrophysics Data System (ADS)

Adib, A.; Afzal, P.; Heydarzadeh, K.

2015-01-01

The aim of this study is to classify the site effect using concentration-area (C-A) fractal model in Meybod city, central Iran, based on microtremor data analysis. Log-log plots of the frequency, amplification and vulnerability index (k-g) indicate a multifractal nature for the parameters in the area. The results obtained from the C-A fractal modelling reveal that proper soil types are located around the central city. The results derived via the fractal modelling were utilized to improve the Nogoshi and Igarashi (1970, 1971) classification results in the Meybod city. The resulting categories are: (1) hard soil and weak rock with frequency of 6.2 to 8 Hz, (2) stiff soil with frequency of about 4.9 to 6.2 Hz, (3) moderately soft soil with the frequency of 2.4 to 4.9 Hz, and (4) soft soil with the frequency lower than 2.4 Hz.

Site effect classification based on microtremor data analysis using concentration-area fractal model

NASA Astrophysics Data System (ADS)

Adib, A.; Afzal, P.; Heydarzadeh, K.

2014-07-01

The aim of this study is to classify the site effect using concentration-area (C-A) fractal model in Meybod city, Central Iran, based on microtremor data analysis. Log-log plots of the frequency, amplification and vulnerability index (k-g) indicate a multifractal nature for the parameters in the area. The results obtained from the C-A fractal modeling reveal that proper soil types are located around the central city. The results derived via the fractal modeling were utilized to improve the Nogoshi's classification results in the Meybod city. The resulted categories are: (1) hard soil and weak rock with frequency of 6.2 to 8 Hz, (2) stiff soil with frequency of about 4.9 to 6.2 Hz, (3) moderately soft soil with the frequency of 2.4 to 4.9 Hz, and (4) soft soil with the frequency lower than 2.4 Hz.

Multi-Site Diagnostic Classification of Schizophrenia Using Discriminant Deep Learning with Functional Connectivity MRI.

PubMed

Zeng, Ling-Li; Wang, Huaning; Hu, Panpan; Yang, Bo; Pu, Weidan; Shen, Hui; Chen, Xingui; Liu, Zhening; Yin, Hong; Tan, Qingrong; Wang, Kai; Hu, Dewen

2018-04-01

A lack of a sufficiently large sample at single sites causes poor generalizability in automatic diagnosis classification of heterogeneous psychiatric disorders such as schizophrenia based on brain imaging scans. Advanced deep learning methods may be capable of learning subtle hidden patterns from high dimensional imaging data, overcome potential site-related variation, and achieve reproducible cross-site classification. However, deep learning-based cross-site transfer classification, despite less imaging site-specificity and more generalizability of diagnostic models, has not been investigated in schizophrenia. A large multi-site functional MRI sample (n = 734, including 357 schizophrenic patients from seven imaging resources) was collected, and a deep discriminant autoencoder network, aimed at learning imaging site-shared functional connectivity features, was developed to discriminate schizophrenic individuals from healthy controls. Accuracies of approximately 85·0% and 81·0% were obtained in multi-site pooling classification and leave-site-out transfer classification, respectively. The learned functional connectivity features revealed dysregulation of the cortical-striatal-cerebellar circuit in schizophrenia, and the most discriminating functional connections were primarily located within and across the default, salience, and control networks. The findings imply that dysfunctional integration of the cortical-striatal-cerebellar circuit across the default, salience, and control networks may play an important role in the "disconnectivity" model underlying the pathophysiology of schizophrenia. The proposed discriminant deep learning method may be capable of learning reliable connectome patterns and help in understanding the pathophysiology and achieving accurate prediction of schizophrenia across multiple independent imaging sites. Copyright © 2018 German Center for Neurodegenerative Diseases (DZNE). Published by Elsevier B.V. All rights reserved.

Kubiëna's heritage: worries and hopes about micropedology (Philippe Duchaufour Medal Lecture)

NASA Astrophysics Data System (ADS)

Stoops, Georges

2010-05-01

Kubiëna's book 'Micropedology' (1938) is considered as the start of soil micromorphology, providing the first concepts allowing a systematic description and comparison of soil thin sections as a central tool for understanding soil genesis and for soil classification. The aim of this contribution is to evaluate the impact and the role of micromorphology in different fields of application, and to evaluate its progress as a discipline. The most important application in soil science has always been in the field of soil genesis. This is however affected by the declining interest (and sponsoring) for genesis nowadays. It remains however a must for studies on pedogenesis and weathering. After a strong impulse early in the nineteen sixties, caused by the study of many exotic soils and the development of new soil classification systems (7th Approximation, later Soil Taxonomy) the role of micromorphology declined together with the general interest in soil classification. Its break through as an instrument in classification did not realise. Several causes can be mentioned. On the base of experience gained in the fields of pedogenesis and classification, micromorphology became for geologists and geomorphologists an important instrument in palaeopedology, Quaternary geology and environmental reconstruction. The last two decades an enormous expansion of micromorphological studies has been noticed in the field of archaeology, not only related to ancient soils, but also to many anthropogenic materials. Archaeologists are probably the most intense users of this discipline now. Since the end of the nineteen sixties quantitative micromorphology (micromorphometry) was developed as a response to the demand for numerical data. It expanded mainly since the development of personal computers, but its wider use is essentially restricted to porosity studies related to soil physics. The complete absence of standardisation of methods and parameters hinders however its use and further expansion. Micromorphology proved also to be precious tool in monitoring experiments, both in the laboratory and in the field, often using quantitative data. Changes become visible in thin sections before they can be detected by other methods. Examples are studies on surface crust formation, effects of freezing, gypsum crystallisation and land management. Last years especially archaeologists contributed in these fields. It is also an excellent tool for controlling and interpreting data obtained by other methods. Analysis of literature and abstracts of congresses show that the last two decennia very few contributions were made related to development of micromorphological concepts and techniques. There are several causes for this situation. The bottleneck hindering use and expansion of micromorphology are both technical and theoretical. The main factors are the difficulty to acquire the necessary basic knowledge of optical techniques and micromorphological interpretation, and the difficulty to prepare good thin sections. Solutions are discussed, even as new opportunities for this discipline, at the benefit of different earth sciences.

Soil evaluation for land use optimizing

NASA Astrophysics Data System (ADS)

Marinina, O. A.

2018-01-01

The article presents the method of soil classification proposed in the course of the study in which the list of indicators proposed by the existing recommendations is optimized. On the example of one of the river basins within the boundaries of the Belgorod region zoning of the territory was carried out. With this approach, the boundaries of the territorial zones are projected along the natural boundaries of natural objects and the productivity of soils is determined as the main criterion for zoning. To assess the territory by soil properties, the features of the soil cover of the river basin were studied and vectorization of the soil variety boundaries was carried out. In the land evaluation essential and useful for the growth of crops macro- and minor-nutrient elements necessary for the growth of crops were included. To compare the soils each of the indicators was translated into relative units. The final score of soil quality is calculated as the mean geometric value of scores from 0 to 100 points for the selected diagnostic features. Through the imposition of results of soil classification and proposed by the concept of basin nature management - land management activities, five zones were identified according to the degree of suitability for use in agriculture.

Classification of Effective Soil Depth by Using Multinomial Logistic Regression Analysis

NASA Astrophysics Data System (ADS)

Chang, C. H.; Chan, H. C.; Chen, B. A.

2016-12-01

Classification of effective soil depth is a task of determining the slopeland utilizable limitation in Taiwan. The "Slopeland Conservation and Utilization Act" categorizes the slopeland into agriculture and husbandry land, land suitable for forestry and land for enhanced conservation according to the factors including average slope, effective soil depth, soil erosion and parental rock. However, sit investigation of the effective soil depth requires a cost-effective field work. This research aimed to classify the effective soil depth by using multinomial logistic regression with the environmental factors. The Wen-Shui Watershed located at the central Taiwan was selected as the study areas. The analysis of multinomial logistic regression is performed by the assistance of a Geographic Information Systems (GIS). The effective soil depth was categorized into four levels including deeper, deep, shallow and shallower. The environmental factors of slope, aspect, digital elevation model (DEM), curvature and normalized difference vegetation index (NDVI) were selected for classifying the soil depth. An Error Matrix was then used to assess the model accuracy. The results showed an overall accuracy of 75%. At the end, a map of effective soil depth was produced to help planners and decision makers in determining the slopeland utilizable limitation in the study areas.

Perspectives on Machine Learning for Classification of Schizotypy Using fMRI Data.

PubMed

Madsen, Kristoffer H; Krohne, Laerke G; Cai, Xin-Lu; Wang, Yi; Chan, Raymond C K

2018-03-15

Functional magnetic resonance imaging is capable of estimating functional activation and connectivity in the human brain, and lately there has been increased interest in the use of these functional modalities combined with machine learning for identification of psychiatric traits. While these methods bear great potential for early diagnosis and better understanding of disease processes, there are wide ranges of processing choices and pitfalls that may severely hamper interpretation and generalization performance unless carefully considered. In this perspective article, we aim to motivate the use of machine learning schizotypy research. To this end, we describe common data processing steps while commenting on best practices and procedures. First, we introduce the important role of schizotypy to motivate the importance of reliable classification, and summarize existing machine learning literature on schizotypy. Then, we describe procedures for extraction of features based on fMRI data, including statistical parametric mapping, parcellation, complex network analysis, and decomposition methods, as well as classification with a special focus on support vector classification and deep learning. We provide more detailed descriptions and software as supplementary material. Finally, we present current challenges in machine learning for classification of schizotypy and comment on future trends and perspectives.

Objects Classification by Learning-Based Visual Saliency Model and Convolutional Neural Network.

PubMed

Li, Na; Zhao, Xinbo; Yang, Yongjia; Zou, Xiaochun

2016-01-01

Humans can easily classify different kinds of objects whereas it is quite difficult for computers. As a hot and difficult problem, objects classification has been receiving extensive interests with broad prospects. Inspired by neuroscience, deep learning concept is proposed. Convolutional neural network (CNN) as one of the methods of deep learning can be used to solve classification problem. But most of deep learning methods, including CNN, all ignore the human visual information processing mechanism when a person is classifying objects. Therefore, in this paper, inspiring the completed processing that humans classify different kinds of objects, we bring forth a new classification method which combines visual attention model and CNN. Firstly, we use the visual attention model to simulate the processing of human visual selection mechanism. Secondly, we use CNN to simulate the processing of how humans select features and extract the local features of those selected areas. Finally, not only does our classification method depend on those local features, but also it adds the human semantic features to classify objects. Our classification method has apparently advantages in biology. Experimental results demonstrated that our method made the efficiency of classification improve significantly.

Observation versus classification in supervised category learning.

PubMed

Levering, Kimery R; Kurtz, Kenneth J

2015-02-01

The traditional supervised classification paradigm encourages learners to acquire only the knowledge needed to predict category membership (a discriminative approach). An alternative that aligns with important aspects of real-world concept formation is learning with a broader focus to acquire knowledge of the internal structure of each category (a generative approach). Our work addresses the impact of a particular component of the traditional classification task: the guess-and-correct cycle. We compare classification learning to a supervised observational learning task in which learners are shown labeled examples but make no classification response. The goals of this work sit at two levels: (1) testing for differences in the nature of the category representations that arise from two basic learning modes; and (2) evaluating the generative/discriminative continuum as a theoretical tool for understand learning modes and their outcomes. Specifically, we view the guess-and-correct cycle as consistent with a more discriminative approach and therefore expected it to lead to narrower category knowledge. Across two experiments, the observational mode led to greater sensitivity to distributional properties of features and correlations between features. We conclude that a relatively subtle procedural difference in supervised category learning substantially impacts what learners come to know about the categories. The results demonstrate the value of the generative/discriminative continuum as a tool for advancing the psychology of category learning and also provide a valuable constraint for formal models and associated theories.

Implicit structured sequence learning: an fMRI study of the structural mere-exposure effect

PubMed Central

Folia, Vasiliki; Petersson, Karl Magnus

2014-01-01

In this event-related fMRI study we investigated the effect of 5 days of implicit acquisition on preference classification by means of an artificial grammar learning (AGL) paradigm based on the structural mere-exposure effect and preference classification using a simple right-linear unification grammar. This allowed us to investigate implicit AGL in a proper learning design by including baseline measurements prior to grammar exposure. After 5 days of implicit acquisition, the fMRI results showed activations in a network of brain regions including the inferior frontal (centered on BA 44/45) and the medial prefrontal regions (centered on BA 8/32). Importantly, and central to this study, the inclusion of a naive preference fMRI baseline measurement allowed us to conclude that these fMRI findings were the intrinsic outcomes of the learning process itself and not a reflection of a preexisting functionality recruited during classification, independent of acquisition. Support for the implicit nature of the knowledge utilized during preference classification on day 5 come from the fact that the basal ganglia, associated with implicit procedural learning, were activated during classification, while the medial temporal lobe system, associated with explicit declarative memory, was consistently deactivated. Thus, preference classification in combination with structural mere-exposure can be used to investigate structural sequence processing (syntax) in unsupervised AGL paradigms with proper learning designs. PMID:24550865

Implicit structured sequence learning: an fMRI study of the structural mere-exposure effect.

PubMed

Folia, Vasiliki; Petersson, Karl Magnus

2014-01-01

In this event-related fMRI study we investigated the effect of 5 days of implicit acquisition on preference classification by means of an artificial grammar learning (AGL) paradigm based on the structural mere-exposure effect and preference classification using a simple right-linear unification grammar. This allowed us to investigate implicit AGL in a proper learning design by including baseline measurements prior to grammar exposure. After 5 days of implicit acquisition, the fMRI results showed activations in a network of brain regions including the inferior frontal (centered on BA 44/45) and the medial prefrontal regions (centered on BA 8/32). Importantly, and central to this study, the inclusion of a naive preference fMRI baseline measurement allowed us to conclude that these fMRI findings were the intrinsic outcomes of the learning process itself and not a reflection of a preexisting functionality recruited during classification, independent of acquisition. Support for the implicit nature of the knowledge utilized during preference classification on day 5 come from the fact that the basal ganglia, associated with implicit procedural learning, were activated during classification, while the medial temporal lobe system, associated with explicit declarative memory, was consistently deactivated. Thus, preference classification in combination with structural mere-exposure can be used to investigate structural sequence processing (syntax) in unsupervised AGL paradigms with proper learning designs.

More than one kind of inference: re-examining what's learned in feature inference and classification.

PubMed

Sweller, Naomi; Hayes, Brett K

2010-08-01

Three studies examined how task demands that impact on attention to typical or atypical category features shape the category representations formed through classification learning and inference learning. During training categories were learned via exemplar classification or by inferring missing exemplar features. In the latter condition inferences were made about missing typical features alone (typical feature inference) or about both missing typical and atypical features (mixed feature inference). Classification and mixed feature inference led to the incorporation of typical and atypical features into category representations, with both kinds of features influencing inferences about familiar (Experiments 1 and 2) and novel (Experiment 3) test items. Those in the typical inference condition focused primarily on typical features. Together with formal modelling, these results challenge previous accounts that have characterized inference learning as producing a focus on typical category features. The results show that two different kinds of inference learning are possible and that these are subserved by different kinds of category representations.

Landmarks of History of Soil Science in Sri Lanka

NASA Astrophysics Data System (ADS)

Mapa, R.

2012-04-01

Sri Lanka is a tropical Island in the Southern tip of Indian subcontinent positioned at 50 55' to 90 50' N latitude and 790 42' to 810 53' E longitude surrounded by the Indian Ocean. It is an island 435 km in length and 224 km width consisting of a land are of 6.56 million ha with a population of 20 million. In area wise it is ranked as 118th in the world, where at present ranked as 47 in population wise and ranked 19th in population density. The country was under colonial rule under Portuguese, Dutch and British from 1505 to 1948. The majority of the people in the past and present earn their living from activities based on land, which indicates the important of the soil resource. The objective of this paper is to describe the landmarks of the history of Soil Science to highlight the achievements and failures, which is useful to enrich our present understanding of Sri Lankan soils. The landmarks of the history of Soil Science in Sri Lanka can be divided to three phases namely, the early period (prior to 1956), the middle period (1956 to 1972) and the present period (from 1972 onwards). During the early period, detailed analytical studies of coffee and tea soils were compiled, and these gave mainly information on up-country soils which led to fertilizer recommendations based on field trials. In addition, rice and forest soils were also studied in less detail. The first classification of Sri Lankan soils and a provisional soil map based on parent material was published by Joachim in 1945 which is a major landmark of history of Soil Science in Sri Lanka. In 1959 Ponnamperuma proposed a soil classification system for wetland rice soils. From 1963 to 1968 valuable information on the land resource was collected and documented by aerial resource surveys funded by Canada-Ceylon Colombo plan aid project. This covered 18 major river basins and about 1/4th of Sri Lanka, which resulted in producing excellent soil maps and information of the areas called the Kelani Aruvi Ara and Walawe basins. The provisional soil map was updated by many other workers as Moorman and Panabokke in 1961 and 1972 using this information. The soil map produced by De Alwis and Panabokke in 1972 at a scale of 1:500,000 was the soil maps mostly used during the past years During the present era, the need for classification of Soils of Sri Lanka according to international methods was felt. A major leap forward in Soil Survey, Classification leading to development of a soil data base was initiated in 1995 with the commencement of the "SRICANSOL" project which was a twining project between the Soil Science Societies of Sri Lanka and Canada. This project is now completed with detail soil maps at a scale of 1:250,000 and soil classified according to international methods for the Wet, Intermediate and Dry zones of Sri Lanka. A digital database consisting of soil profile description and physical and chemical data is under preparation for 28, 40 and 51 benchmark sites of the Wet, Intermediate and Dry zones respectively. The emphases on studies on Soil Science in the country at present is more towards environmental conservation related to soil erosion control, reducing of pollution of soil and water bodies from nitrates, pesticide residues and heavy metal accumulation. Key words: Sri Lanka, Provisional soil map

Comparing the performance of various digital soil mapping approaches to map physical soil properties

NASA Astrophysics Data System (ADS)

Laborczi, Annamária; Takács, Katalin; Pásztor, László

2015-04-01

Spatial information on physical soil properties is intensely expected, in order to support environmental related and land use management decisions. One of the most widely used properties to characterize soils physically is particle size distribution (PSD), which determines soil water management and cultivability. According to their size, different particles can be categorized as clay, silt, or sand. The size intervals are defined by national or international textural classification systems. The relative percentage of sand, silt, and clay in the soil constitutes textural classes, which are also specified miscellaneously in various national and/or specialty systems. The most commonly used is the classification system of the United States Department of Agriculture (USDA). Soil texture information is essential input data in meteorological, hydrological and agricultural prediction modelling. Although Hungary has a great deal of legacy soil maps and other relevant soil information, it often occurs, that maps do not exist on a certain characteristic with the required thematic and/or spatial representation. The recent developments in digital soil mapping (DSM), however, provide wide opportunities for the elaboration of object specific soil maps (OSSM) with predefined parameters (resolution, accuracy, reliability etc.). Due to the simultaneous richness of available Hungarian legacy soil data, spatial inference methods and auxiliary environmental information, there is a high versatility of possible approaches for the compilation of a given soil map. This suggests the opportunity of optimization. For the creation of an OSSM one might intend to identify the optimum set of soil data, method and auxiliary co-variables optimized for the resources (data costs, computation requirements etc.). We started comprehensive analysis of the effects of the various DSM components on the accuracy of the output maps on pilot areas. The aim of this study is to compare and evaluate different digital soil mapping methods and sets of ancillary variables for producing the most accurate spatial prediction of texture classes in a given area of interest. Both legacy and recently collected data on PSD were used as reference information. The predictor variable data set consisted of digital elevation model and its derivatives, lithology, land use maps as well as various bands and indices of satellite images. Two conceptionally different approaches can be applied in the mapping process. Textural classification can be realized after particle size data were spatially extended by proper geostatistical method. Alternatively, the textural classification is carried out first, followed by the spatial extension through suitable data mining method. According to the first approach, maps of sand, silt and clay percentage have been computed through regression kriging (RK). Since the three maps are compositional (their sum must be 100%), we applied Additive Log-Ratio (alr) transformation, instead of kriging them independently. Finally, the texture class map has been compiled according to the USDA categories from the three maps. Different combinations of reference and training soil data and auxiliary covariables resulted several different maps. On the basis of the other way, the PSD were classified firstly into the USDA categories, then the texture class maps were compiled directly by data mining methods (classification trees and random forests). The various results were compared to each other as well as to the RK maps. The performance of the different methods and data sets has been examined by testing the accuracy of the geostatistically computed and the directly classified results to assess the most predictive and accurate method. Acknowledgement: Our work was supported by the Hungarian National Scientific Research Foundation (OTKA, Grant No. K105167).

Soil Science in Space: Thinking Way Outside the Box

NASA Technical Reports Server (NTRS)

Ming, D. W.

2016-01-01

Mars is a perfect laboratory to reconsider the future of pedology across the universe. By investigating the soils and geology through our Curiosity and further endeavors, we find ourselves able to learn about the past, present, and possibly the future. Imagine what we could learn about the early Earth if we could have explored it without vegetation and clouds in the way. The tools and techniques that are used to probe the Martian soil can teach us about exploring the soils on Earth. Although many may feel that soil science has learned all that it can about the soils on Earth, we know differently. Deciding what the most important things to know about Martian soils can help us focus on the fundamentals of soil science on Earth. Our soil science knowledge and experience on Earth can help us learn more about the angry red planet. Why is it so angry with so many fascinating secrets it can tell?

An ecological classification system for the central hardwoods region: The Hoosier National Forest

Treesearch

James E. Van Kley; George R. Parker

1993-01-01

This study, a multifactor ecological classification system, using vegetation, soil characteristics, and physiography, was developed for the landscape of the Hoosier National Forest in Southern Indiana. Measurements of ground flora, saplings, and canopy trees from selected stands older than 80 years were subjected to TWINSPAN classification and DECORANA ordination....

Seismic Site Classification and Empirical Correlation Between Standard Penetration Test N Value and Shear Wave Velocity for Guwahati Based on Thorough Subsoil Investigation Data

NASA Astrophysics Data System (ADS)

Kumar, Abhishek; Harinarayan, N. H.; Verma, Vishal; Anand, Saurabh; Borah, Uddipana; Bania, Mousumi

2018-04-01

Guwahati, the Gateway of India in the northeast, is a large business and development center. Past seismic scenarios suggest moderate to significant effects of regional earthquakes (EQs) in Guwahati in terms of liquefaction as well as building damages. Considering the role of local soil in amplifying EQ-generated ground motions and controlling surface damages, present study attempts seismic site classification of subsoil of Guwahati. Subsoil is explored based on 43 geophysical tests and 244 borelogs gathered from different resources. Based on the borehole data, 4 numbers of 2D cross-sections are developed from different parts of Guwahati, clearly indicating that a majority of the locations are composed of clay of intermediate to high plasticity while at specific locations only, layers of sand are found at selective depths. Further, seismic site classification based on 30 m average SPT-N suggests that a major part of Guwahati falls under seismic site class (SSC) D such as Balaji Temple and Airport. However, Assam Zoo, Pan Bazaar, IIT campus, Dhol Gobinda and Maligaon show SSC E clearly indicating the presence of soft soil deposits at these locations. Similar site classification is also attempted from MASW test-based 30 m average shear wave velocity (V S30). V S30-based site classification also categorizes most of Guwahati under SSC D. However, there are locations in the southern part of Guwahati which belong to SSC C as well. Mismatch in SSC based on two different test findings for Indian soil found here are consistent with previous studies. Further, three empirical correlations based on both SPT-N and V S profiles at 22 test locations are developed for: (1) clayey; (2) sandy and (3) all soil types. Proposed correlation for all soil types is validated graphically and is found closely matching with similar correlations for Turkey and Lucknow.

Mapping soil features from multispectral scanner data

NASA Technical Reports Server (NTRS)

Kristof, S. J.; Zachary, A. L.

1974-01-01

In being able to identify quickly gross variations in soil features, the computer-aided classification of multispectral scanner data can be an effective aid to soil surveying. Variations in soil tone are easily seen as well as variations in features related to soil tone, e.g., drainage patterns and organic matter content. Changes in surface texture also affect the reflectance properties of soils. Inasmuch as conventional soil classes are based on both surface and subsurface soil characteristics, the technique described here can be expected only to augment and not replace traditional soil mapping.

Deep Multi-Task Learning for Tree Genera Classification

NASA Astrophysics Data System (ADS)

Ko, C.; Kang, J.; Sohn, G.

2018-05-01

The goal for our paper is to classify tree genera using airborne Light Detection and Ranging (LiDAR) data with Convolution Neural Network (CNN) - Multi-task Network (MTN) implementation. Unlike Single-task Network (STN) where only one task is assigned to the learning outcome, MTN is a deep learning architect for learning a main task (classification of tree genera) with other tasks (in our study, classification of coniferous and deciduous) simultaneously, with shared classification features. The main contribution of this paper is to improve classification accuracy from CNN-STN to CNN-MTN. This is achieved by introducing a concurrence loss (Lcd) to the designed MTN. This term regulates the overall network performance by minimizing the inconsistencies between the two tasks. Results show that we can increase the classification accuracy from 88.7 % to 91.0 % (from STN to MTN). The second goal of this paper is to solve the problem of small training sample size by multiple-view data generation. The motivation of this goal is to address one of the most common problems in implementing deep learning architecture, the insufficient number of training data. We address this problem by simulating training dataset with multiple-view approach. The promising results from this paper are providing a basis for classifying a larger number of dataset and number of classes in the future.

Semi-supervised learning for ordinal Kernel Discriminant Analysis.

PubMed

Pérez-Ortiz, M; Gutiérrez, P A; Carbonero-Ruz, M; Hervás-Martínez, C

2016-12-01

Ordinal classification considers those classification problems where the labels of the variable to predict follow a given order. Naturally, labelled data is scarce or difficult to obtain in this type of problems because, in many cases, ordinal labels are given by a user or expert (e.g. in recommendation systems). Firstly, this paper develops a new strategy for ordinal classification where both labelled and unlabelled data are used in the model construction step (a scheme which is referred to as semi-supervised learning). More specifically, the ordinal version of kernel discriminant learning is extended for this setting considering the neighbourhood information of unlabelled data, which is proposed to be computed in the feature space induced by the kernel function. Secondly, a new method for semi-supervised kernel learning is devised in the context of ordinal classification, which is combined with our developed classification strategy to optimise the kernel parameters. The experiments conducted compare 6 different approaches for semi-supervised learning in the context of ordinal classification in a battery of 30 datasets, showing (1) the good synergy of the ordinal version of discriminant analysis and the use of unlabelled data and (2) the advantage of computing distances in the feature space induced by the kernel function. Copyright © 2016 Elsevier Ltd. All rights reserved.

Multi-level discriminative dictionary learning with application to large scale image classification.

PubMed

Shen, Li; Sun, Gang; Huang, Qingming; Wang, Shuhui; Lin, Zhouchen; Wu, Enhua

2015-10-01

The sparse coding technique has shown flexibility and capability in image representation and analysis. It is a powerful tool in many visual applications. Some recent work has shown that incorporating the properties of task (such as discrimination for classification task) into dictionary learning is effective for improving the accuracy. However, the traditional supervised dictionary learning methods suffer from high computation complexity when dealing with large number of categories, making them less satisfactory in large scale applications. In this paper, we propose a novel multi-level discriminative dictionary learning method and apply it to large scale image classification. Our method takes advantage of hierarchical category correlation to encode multi-level discriminative information. Each internal node of the category hierarchy is associated with a discriminative dictionary and a classification model. The dictionaries at different layers are learnt to capture the information of different scales. Moreover, each node at lower layers also inherits the dictionary of its parent, so that the categories at lower layers can be described with multi-scale information. The learning of dictionaries and associated classification models is jointly conducted by minimizing an overall tree loss. The experimental results on challenging data sets demonstrate that our approach achieves excellent accuracy and competitive computation cost compared with other sparse coding methods for large scale image classification.

Appendix E: Representative Pedon Descriptions for the Soils of the GLEES Wyoming Soil Survey Area: EGL, WGL & Lost Lake Watersheds

Treesearch

R. W. E. Hopper; P. M. Walthall

1994-01-01

A soil survey was conducted of the East Glacier, West Glacier and Lost Lake watersheds in July-September 1986. Procedures appropriate for an Order 3 soil survey were followed. Fifteen locations were surveyed and a total of 166 samples were analyzed. The 15 series are listed below, along with the soil classification and number of samples analyzed.

The Effect of Soil Properties on Metal Bioavailability: Field Scale Validation to Support Regulatory Acceptance

DTIC Science & Technology

2013-06-01

Bioavailability, metals, soil, bioaccessibility, ecological risk, arsenic, cadmium , chromium, lead 16. SECURITY CLASSIFICATION OF:U 17. LIMITATION...located in Sacramento, CA. Soils from a former wastewater treatment lagoon are contaminated with high concentrations of lead , chromium, and cadmium ...in soil. Soil and Sediment Contamination, 2003. 12(1): p. 1-21. 23. Pierzynski, G.M. and A.P. Schwab, Bioavailability of Zinc, Cadmium , and Lead

Classification of andisol soil on robusta coffee plantation in Silima Pungga - Pungga District

NASA Astrophysics Data System (ADS)

Marbun, P.; Nasution, Z.; Hanum, H.; Karim, A.

2018-02-01

The survey study aims to classify the Inceptisol soil on Robusta coffee plantation in Silima Pugga-Pungga District, from Order level to Sub Group level. The study was conducted on location of sample soil profiles which were determined based on Soil Map Unit (SMU) with the main Andisol Order, i.e. SMU 12, SMU 15 and SMU 17 of 18 existing SMU. The soil profiles were described to determine the morphological characteristics of the soil, while the physical and chemical properties were done by laboratory analysis. The soil samples were taken from each horizon in each profile and analyzed in the laboratory in the form of soil texture, bulk density, pH H2O, pH KCl, pH NaF, C-organic, exchangeable bases (Ca2+, Mg2+, K+, Na+), ZPC (zero point charge), base saturation, cation exchange capasity (CEC), P-retention, Al-Oxalate (Al-O) and Si-Oxalate (Si-O). The results showed that the classification of Andisol soil based on Soil Taxonomy only has one Sub Group namely Typic Hapludand. It is expected that the results of this study can provide information for more appropriate land management in order to increase the production of Robusta coffee plant in Silima Pungga-Pungga Sub district.

Weakly supervised classification in high energy physics

DOE PAGES

Dery, Lucio Mwinmaarong; Nachman, Benjamin; Rubbo, Francesco; ...

2017-05-01

As machine learning algorithms become increasingly sophisticated to exploit subtle features of the data, they often become more dependent on simulations. Here, this paper presents a new approach called weakly supervised classification in which class proportions are the only input into the machine learning algorithm. Using one of the most challenging binary classification tasks in high energy physics $-$ quark versus gluon tagging $-$ we show that weakly supervised classification can match the performance of fully supervised algorithms. Furthermore, by design, the new algorithm is insensitive to any mis-modeling of discriminating features in the data by the simulation. Weakly supervisedmore » classification is a general procedure that can be applied to a wide variety of learning problems to boost performance and robustness when detailed simulations are not reliable or not available.« less

Weakly supervised classification in high energy physics

DOE Office of Scientific and Technical Information (OSTI.GOV)

Dery, Lucio Mwinmaarong; Nachman, Benjamin; Rubbo, Francesco

As machine learning algorithms become increasingly sophisticated to exploit subtle features of the data, they often become more dependent on simulations. Here, this paper presents a new approach called weakly supervised classification in which class proportions are the only input into the machine learning algorithm. Using one of the most challenging binary classification tasks in high energy physics $-$ quark versus gluon tagging $-$ we show that weakly supervised classification can match the performance of fully supervised algorithms. Furthermore, by design, the new algorithm is insensitive to any mis-modeling of discriminating features in the data by the simulation. Weakly supervisedmore » classification is a general procedure that can be applied to a wide variety of learning problems to boost performance and robustness when detailed simulations are not reliable or not available.« less

Model-based classification of CPT data and automated lithostratigraphic mapping for high-resolution characterization of a heterogeneous sedimentary aquifer

PubMed Central

Mallants, Dirk; Batelaan, Okke; Gedeon, Matej; Huysmans, Marijke; Dassargues, Alain

2017-01-01

Cone penetration testing (CPT) is one of the most efficient and versatile methods currently available for geotechnical, lithostratigraphic and hydrogeological site characterization. Currently available methods for soil behaviour type classification (SBT) of CPT data however have severe limitations, often restricting their application to a local scale. For parameterization of regional groundwater flow or geotechnical models, and delineation of regional hydro- or lithostratigraphy, regional SBT classification would be very useful. This paper investigates the use of model-based clustering for SBT classification, and the influence of different clustering approaches on the properties and spatial distribution of the obtained soil classes. We additionally propose a methodology for automated lithostratigraphic mapping of regionally occurring sedimentary units using SBT classification. The methodology is applied to a large CPT dataset, covering a groundwater basin of ~60 km2 with predominantly unconsolidated sandy sediments in northern Belgium. Results show that the model-based approach is superior in detecting the true lithological classes when compared to more frequently applied unsupervised classification approaches or literature classification diagrams. We demonstrate that automated mapping of lithostratigraphic units using advanced SBT classification techniques can provide a large gain in efficiency, compared to more time-consuming manual approaches and yields at least equally accurate results. PMID:28467468

Model-based classification of CPT data and automated lithostratigraphic mapping for high-resolution characterization of a heterogeneous sedimentary aquifer.

PubMed

Rogiers, Bart; Mallants, Dirk; Batelaan, Okke; Gedeon, Matej; Huysmans, Marijke; Dassargues, Alain

2017-01-01

Cone penetration testing (CPT) is one of the most efficient and versatile methods currently available for geotechnical, lithostratigraphic and hydrogeological site characterization. Currently available methods for soil behaviour type classification (SBT) of CPT data however have severe limitations, often restricting their application to a local scale. For parameterization of regional groundwater flow or geotechnical models, and delineation of regional hydro- or lithostratigraphy, regional SBT classification would be very useful. This paper investigates the use of model-based clustering for SBT classification, and the influence of different clustering approaches on the properties and spatial distribution of the obtained soil classes. We additionally propose a methodology for automated lithostratigraphic mapping of regionally occurring sedimentary units using SBT classification. The methodology is applied to a large CPT dataset, covering a groundwater basin of ~60 km2 with predominantly unconsolidated sandy sediments in northern Belgium. Results show that the model-based approach is superior in detecting the true lithological classes when compared to more frequently applied unsupervised classification approaches or literature classification diagrams. We demonstrate that automated mapping of lithostratigraphic units using advanced SBT classification techniques can provide a large gain in efficiency, compared to more time-consuming manual approaches and yields at least equally accurate results.

Argumentation Based Joint Learning: A Novel Ensemble Learning Approach

PubMed Central

Xu, Junyi; Yao, Li; Li, Le

2015-01-01

Recently, ensemble learning methods have been widely used to improve classification performance in machine learning. In this paper, we present a novel ensemble learning method: argumentation based multi-agent joint learning (AMAJL), which integrates ideas from multi-agent argumentation, ensemble learning, and association rule mining. In AMAJL, argumentation technology is introduced as an ensemble strategy to integrate multiple base classifiers and generate a high performance ensemble classifier. We design an argumentation framework named Arena as a communication platform for knowledge integration. Through argumentation based joint learning, high quality individual knowledge can be extracted, and thus a refined global knowledge base can be generated and used independently for classification. We perform numerous experiments on multiple public datasets using AMAJL and other benchmark methods. The results demonstrate that our method can effectively extract high quality knowledge for ensemble classifier and improve the performance of classification. PMID:25966359

QUEST: Eliminating Online Supervised Learning for Efficient Classification Algorithms.

PubMed

Zwartjes, Ardjan; Havinga, Paul J M; Smit, Gerard J M; Hurink, Johann L

2016-10-01

In this work, we introduce QUEST (QUantile Estimation after Supervised Training), an adaptive classification algorithm for Wireless Sensor Networks (WSNs) that eliminates the necessity for online supervised learning. Online processing is important for many sensor network applications. Transmitting raw sensor data puts high demands on the battery, reducing network life time. By merely transmitting partial results or classifications based on the sampled data, the amount of traffic on the network can be significantly reduced. Such classifications can be made by learning based algorithms using sampled data. An important issue, however, is the training phase of these learning based algorithms. Training a deployed sensor network requires a lot of communication and an impractical amount of human involvement. QUEST is a hybrid algorithm that combines supervised learning in a controlled environment with unsupervised learning on the location of deployment. Using the SITEX02 dataset, we demonstrate that the presented solution works with a performance penalty of less than 10% in 90% of the tests. Under some circumstances, it even outperforms a network of classifiers completely trained with supervised learning. As a result, the need for on-site supervised learning and communication for training is completely eliminated by our solution.

7 CFR 601.1 - Functions assigned.

Code of Federal Regulations, 2013 CFR

2013-01-01

...) National Resources Inventory (NRI) that is a statistically-based survey designed and implemented using... both the provisions of the Food Security Act and Section 404 of the Clean Water Act. (ii) Soil surveys.... Soil surveys are based on scientific analysis and classification of the soils and are used to determine...

7 CFR 601.1 - Functions assigned.

Code of Federal Regulations, 2011 CFR

2011-01-01

...) National Resources Inventory (NRI) that is a statistically-based survey designed and implemented using... both the provisions of the Food Security Act and Section 404 of the Clean Water Act. (ii) Soil surveys.... Soil surveys are based on scientific analysis and classification of the soils and are used to determine...

7 CFR 601.1 - Functions assigned.

Code of Federal Regulations, 2012 CFR

2012-01-01

...) National Resources Inventory (NRI) that is a statistically-based survey designed and implemented using... both the provisions of the Food Security Act and Section 404 of the Clean Water Act. (ii) Soil surveys.... Soil surveys are based on scientific analysis and classification of the soils and are used to determine...

7 CFR 601.1 - Functions assigned.

Code of Federal Regulations, 2014 CFR

2014-01-01

...) National Resources Inventory (NRI) that is a statistically-based survey designed and implemented using... both the provisions of the Food Security Act and Section 404 of the Clean Water Act. (ii) Soil surveys.... Soil surveys are based on scientific analysis and classification of the soils and are used to determine...

7 CFR 601.1 - Functions assigned.

Code of Federal Regulations, 2010 CFR

2010-01-01

...) National Resources Inventory (NRI) that is a statistically-based survey designed and implemented using... both the provisions of the Food Security Act and Section 404 of the Clean Water Act. (ii) Soil surveys.... Soil surveys are based on scientific analysis and classification of the soils and are used to determine...

29 CFR Appendix A to Subpart P of... - Soil Classification

Code of Federal Regulations, 2010 CFR

2010-07-01

... the particles are held together by a chemical agent, such as calcium carbonate, such that a hand-size sample cannot be crushed into powder or individual soil particles by finger pressure. Cohesive soil means... quantitative and qualitative information as may be necessary to identify properly the properties, factors, and...

Soils of northern spurs of the Cherskii Ridge in the area of the northern pole of cold: Morphology, properties, and classification

NASA Astrophysics Data System (ADS)

Okoneshnikova, M. V.; Desyatkin, R. V.

2017-08-01

The soils in the area of the northern pole of cold located on the interfluve between the Yana and Adycha rivers within the spurs of Kisilyakh Ridge included in the mountain system of Cherskii Ridge have been studied for the first time. The profile-genetic approach has been applied to describe the soils and determine their classification position. It is found that the major soil types in this region are the soils of the postlithogenic trunk belonging to the orders of lithozems (Cryic Leptosols), gley soils (Gleyic Skeletic Cryosols), and Al-Fe-humus soils (Spodic Skeletic Cryosols). The ecological ranges of altitudinal zones— the taiga zone with various types of lithozems below 630-700 m a.s.l. and the tundra zone with combinations of gley and nongley cryogenic soils above these heights—have been established. The development of gley or nongley soils is specified by the local orogenic and lithological conditions and slope aspect, which, in turn, control the degree of drainage and the presence and character of permafrost. In the profile of mountainous gley soils (gleyzems) with shallow ice-rich permafrost, cryogenic processes and features typical of the analogues of these soils on plains—cryogenic cracking, cryoturbation, solifluction, thixotropy, oxiaquic features above permafrost, saturation of the soil profile with mobile humus, etc.—are typical.

An Illustration of Diagnostic Classification Modeling in Student Learning Outcomes Assessment

ERIC Educational Resources Information Center

Jurich, Daniel P.; Bradshaw, Laine P.

2014-01-01

The assessment of higher-education student learning outcomes is an important component in understanding the strengths and weaknesses of academic and general education programs. This study illustrates the application of diagnostic classification models, a burgeoning set of statistical models, in assessing student learning outcomes. To facilitate…

Analyzing Student Inquiry Data Using Process Discovery and Sequence Classification

ERIC Educational Resources Information Center

Emond, Bruno; Buffett, Scott

2015-01-01

This paper reports on results of applying process discovery mining and sequence classification mining techniques to a data set of semi-structured learning activities. The main research objective is to advance educational data mining to model and support self-regulated learning in heterogeneous environments of learning content, activities, and…

29 CFR Appendix B to Subpart P of... - Sloping and Benching

Code of Federal Regulations, 2013 CFR

2013-07-01

... excavations 20 feet or less in depth made in layered soils shall have a maximum allowable slope for each layer.... Distress means that the soil is in a condition where a cave-in is imminent or is likely to occur. Distress... 24 hours that an excavation is open. (c) Requirements—(1) Soil classification. Soil and rock deposits...

29 CFR Appendix B to Subpart P of... - Sloping and Benching

Code of Federal Regulations, 2012 CFR

2012-07-01

... excavations 20 feet or less in depth made in layered soils shall have a maximum allowable slope for each layer.... Distress means that the soil is in a condition where a cave-in is imminent or is likely to occur. Distress... 24 hours that an excavation is open. (c) Requirements—(1) Soil classification. Soil and rock deposits...

29 CFR Appendix B to Subpart P of... - Sloping and Benching

Code of Federal Regulations, 2014 CFR

2014-07-01

... excavations 20 feet or less in depth made in layered soils shall have a maximum allowable slope for each layer.... Distress means that the soil is in a condition where a cave-in is imminent or is likely to occur. Distress... 24 hours that an excavation is open. (c) Requirements—(1) Soil classification. Soil and rock deposits...

29 CFR Appendix B to Subpart P of... - Sloping and Benching

Code of Federal Regulations, 2011 CFR

2011-07-01

... excavations 20 feet or less in depth made in layered soils shall have a maximum allowable slope for each layer.... Distress means that the soil is in a condition where a cave-in is imminent or is likely to occur. Distress... 24 hours that an excavation is open. (c) Requirements—(1) Soil classification. Soil and rock deposits...

29 CFR Appendix B to Subpart P of... - Sloping and Benching

Code of Federal Regulations, 2010 CFR

2010-07-01

... excavations 20 feet or less in depth made in layered soils shall have a maximum allowable slope for each layer.... Distress means that the soil is in a condition where a cave-in is imminent or is likely to occur. Distress... 24 hours that an excavation is open. (c) Requirements—(1) Soil classification. Soil and rock deposits...

Oscillatory neural network for pattern recognition: trajectory based classification and supervised learning.

PubMed

Miller, Vonda H; Jansen, Ben H

2008-12-01

Computer algorithms that match human performance in recognizing written text or spoken conversation remain elusive. The reasons why the human brain far exceeds any existing recognition scheme to date in the ability to generalize and to extract invariant characteristics relevant to category matching are not clear. However, it has been postulated that the dynamic distribution of brain activity (spatiotemporal activation patterns) is the mechanism by which stimuli are encoded and matched to categories. This research focuses on supervised learning using a trajectory based distance metric for category discrimination in an oscillatory neural network model. Classification is accomplished using a trajectory based distance metric. Since the distance metric is differentiable, a supervised learning algorithm based on gradient descent is demonstrated. Classification of spatiotemporal frequency transitions and their relation to a priori assessed categories is shown along with the improved classification results after supervised training. The results indicate that this spatiotemporal representation of stimuli and the associated distance metric is useful for simple pattern recognition tasks and that supervised learning improves classification results.

Deep learning application: rubbish classification with aid of an android device

NASA Astrophysics Data System (ADS)

Liu, Sijiang; Jiang, Bo; Zhan, Jie

2017-06-01

Deep learning is a very hot topic currently in pattern recognition and artificial intelligence researches. Aiming at the practical problem that people usually don't know correct classifications some rubbish should belong to, based on the powerful image classification ability of the deep learning method, we have designed a prototype system to help users to classify kinds of rubbish. Firstly the CaffeNet Model was adopted for our classification network training on the ImageNet dataset, and the trained network was deployed on a web server. Secondly an android app was developed for users to capture images of unclassified rubbish, upload images to the web server for analyzing backstage and retrieve the feedback, so that users can obtain the classification guide by an android device conveniently. Tests on our prototype system of rubbish classification show that: an image of one single type of rubbish with origin shape can be better used to judge its classification, while an image containing kinds of rubbish or rubbish with changed shape may fail to help users to decide rubbish's classification. However, the system still shows promising auxiliary function for rubbish classification if the network training strategy can be optimized further.

Soil classification for seismic site effect using MASW and ReMi methods: A case study from western Anatolia (Dikili -İzmir)

NASA Astrophysics Data System (ADS)

Karabulut, Savaş

2018-03-01

The study area is located in the northern part of Izmir, Western Turkey, prone to an active tectonic extensional regime and includes typical features of sedimentary basins, horst-grabens surrounded by a series of normal and strike-slip faults. In September 1939 the Dikili (Kabakum) earthquake with a magnitude of Mw: 6.6 occurred and after this phenomenon, residents moved from the west of Dikili to the east (i.d. soft sediments to relative to rock area). A proper estimate of the earthquake-related hazard for the area is the main objective of this study. The site effect and soil engineering problems for estimating hazard parameters at the soil surface need to be carefully analyzed for seismic site classification and geo-engineering problems like soil liquefaction, soil settlement, soil bearing capacity and soil amplification. To solve the soil static and dynamic problems, shear-wave velocities have been used in a joint interpretation process; Multichannel Analysis of Surface Waves (MASW) and Refraction Microtremor (ReMi) analyses were conducted on 121 sites with 300 × 300 m grid size in an area of 60 km2. It has been proposed that the probability of an earthquake with a magnitude of Mw: 6 occurring within 10 years is 64%, when considering the Gutenberg-Richter model. This puts the region under an important earthquake risk. The estimated Vs30 values are ≤180 m/s in the central and the northernmost part of the study area are showing an E type soil after the classification of NEHRP, where alluvial deposits are dominant. Vs30 values in the north and central part are between 180 ≤ Vs ≤ 360 m/s suggesting a D type soil. In the southernmost part of the study area where volcanic rocks are widely distributed, Vs30 values range between 360 and 908 m/s, corresponding to a C type and B type soil. The results show that soil liquefaction induced settlement and soil amplification are the most important problems in the south and the northernmost part of the study area, which is densely populated and encompasses the urbanized part of the study region.

Automatic classification of protein structures using physicochemical parameters.

PubMed

Mohan, Abhilash; Rao, M Divya; Sunderrajan, Shruthi; Pennathur, Gautam

2014-09-01

Protein classification is the first step to functional annotation; SCOP and Pfam databases are currently the most relevant protein classification schemes. However, the disproportion in the number of three dimensional (3D) protein structures generated versus their classification into relevant superfamilies/families emphasizes the need for automated classification schemes. Predicting function of novel proteins based on sequence information alone has proven to be a major challenge. The present study focuses on the use of physicochemical parameters in conjunction with machine learning algorithms (Naive Bayes, Decision Trees, Random Forest and Support Vector Machines) to classify proteins into their respective SCOP superfamily/Pfam family, using sequence derived information. Spectrophores™, a 1D descriptor of the 3D molecular field surrounding a structure was used as a benchmark to compare the performance of the physicochemical parameters. The machine learning algorithms were modified to select features based on information gain for each SCOP superfamily/Pfam family. The effect of combining physicochemical parameters and spectrophores on classification accuracy (CA) was studied. Machine learning algorithms trained with the physicochemical parameters consistently classified SCOP superfamilies and Pfam families with a classification accuracy above 90%, while spectrophores performed with a CA of around 85%. Feature selection improved classification accuracy for both physicochemical parameters and spectrophores based machine learning algorithms. Combining both attributes resulted in a marginal loss of performance. Physicochemical parameters were able to classify proteins from both schemes with classification accuracy ranging from 90-96%. These results suggest the usefulness of this method in classifying proteins from amino acid sequences.

A unified classification model for modeling of seismic liquefaction potential of soil based on CPT

PubMed Central

Samui, Pijush; Hariharan, R.

2014-01-01

The evaluation of liquefaction potential of soil due to an earthquake is an important step in geosciences. This article examines the capability of Minimax Probability Machine (MPM) for the prediction of seismic liquefaction potential of soil based on the Cone Penetration Test (CPT) data. The dataset has been taken from Chi–Chi earthquake. MPM is developed based on the use of hyperplanes. It has been adopted as a classification tool. This article uses two models (MODEL I and MODEL II). MODEL I employs Cone Resistance (qc) and Cyclic Stress Ratio (CSR) as input variables. qc and Peak Ground Acceleration (PGA) have been taken as inputs for MODEL II. The developed MPM gives 100% accuracy. The results show that the developed MPM can predict liquefaction potential of soil based on qc and PGA. PMID:26199749

A unified classification model for modeling of seismic liquefaction potential of soil based on CPT.

PubMed

Samui, Pijush; Hariharan, R

2015-07-01

The evaluation of liquefaction potential of soil due to an earthquake is an important step in geosciences. This article examines the capability of Minimax Probability Machine (MPM) for the prediction of seismic liquefaction potential of soil based on the Cone Penetration Test (CPT) data. The dataset has been taken from Chi-Chi earthquake. MPM is developed based on the use of hyperplanes. It has been adopted as a classification tool. This article uses two models (MODEL I and MODEL II). MODEL I employs Cone Resistance (q c) and Cyclic Stress Ratio (CSR) as input variables. q c and Peak Ground Acceleration (PGA) have been taken as inputs for MODEL II. The developed MPM gives 100% accuracy. The results show that the developed MPM can predict liquefaction potential of soil based on q c and PGA.

Soil Salinity Mapping in Everglades National Park Using Remote Sensing Techniques

NASA Astrophysics Data System (ADS)

Su, H.; Khadim, F. K.; Blankenship, J.; Sobhan, K.

2017-12-01

The South Florida Everglades is a vast subtropical wetland with a globally unique hydrology and ecology, and it is designated as an International Biosphere Reserve and a Wetland of International Importance. Everglades National Park (ENP) is a hydro-ecologically enriched wetland with varying salinity contents, which is a concern for terrestrial ecosystem balance and sustainability. As such, in this study, time series soil salinity mapping was carried out for the ENP area. The mapping first entailed a maximum likelihood classification of seven land cover classes for the ENP area—namely mangrove forest, mangrove scrub, low-density forest, sawgrass, prairies and marshes, barren lands with woodland hammock and water—for the years 1996, 2000, 2006, 2010 and 2015. The classifications for 1996-2010 yielded accuracies of 82%-94%, and the 2015 classification was supported through ground truthing. Afterwards, electric conductivity (EC) tolerance thresholds for each vegetation class were established,which yielded soil salinity maps comprising four soil salinity classes—i.e., the non- (EC = 0 2 dS/m), low- (EC = 2 4 dS/m), moderate- (EC = 4 8 dS/m) and high-saline (EC = >8 dS/m) areas. The soil salinity maps visualized the spatial distribution of soil salinity with no significant temporal variations. The innovative approach of "land cover identification to salinity estimation" used in the study is pragmatic and application oriented, and the study upshots are also useful, considering the diversifying ecological context of the ENP area.

Classification of multiple sclerosis lesions using adaptive dictionary learning.

PubMed

Deshpande, Hrishikesh; Maurel, Pierre; Barillot, Christian

2015-12-01

This paper presents a sparse representation and an adaptive dictionary learning based method for automated classification of multiple sclerosis (MS) lesions in magnetic resonance (MR) images. Manual delineation of MS lesions is a time-consuming task, requiring neuroradiology experts to analyze huge volume of MR data. This, in addition to the high intra- and inter-observer variability necessitates the requirement of automated MS lesion classification methods. Among many image representation models and classification methods that can be used for such purpose, we investigate the use of sparse modeling. In the recent years, sparse representation has evolved as a tool in modeling data using a few basis elements of an over-complete dictionary and has found applications in many image processing tasks including classification. We propose a supervised classification approach by learning dictionaries specific to the lesions and individual healthy brain tissues, which include white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF). The size of the dictionaries learned for each class plays a major role in data representation but it is an even more crucial element in the case of competitive classification. Our approach adapts the size of the dictionary for each class, depending on the complexity of the underlying data. The algorithm is validated using 52 multi-sequence MR images acquired from 13 MS patients. The results demonstrate the effectiveness of our approach in MS lesion classification. Copyright © 2015 Elsevier Ltd. All rights reserved.

Classification and evaluation for forest sites on the Mid-Cumberland Plateau

Treesearch

Glendon W. Smalley

1982-01-01

Presents a comprehensive forest site classification system for the central portion of the Cumberland Plateau in northeast Alabama, and east-central Tennessee. The system is based on physiography, geology, soils, topography, and vegetation.

[Object-oriented remote sensing image classification in epidemiological studies of visceral leishmaniasis in urban areas].

PubMed

Almeida, Andréa Sobral de; Werneck, Guilherme Loureiro; Resendes, Ana Paula da Costa

2014-08-01

This study explored the use of object-oriented classification of remote sensing imagery in epidemiological studies of visceral leishmaniasis (VL) in urban areas. To obtain temperature and environmental information, an object-oriented classification approach was applied to Landsat 5 TM scenes from the city of Teresina, Piauí State, Brazil. For 1993-1996, VL incidence rates correlated positively with census tracts covered by dense vegetation, grass/pasture, and bare soil and negatively with areas covered by water and densely populated areas. In 2001-2006, positive correlations were found with dense vegetation, grass/pasture, bare soil, and densely populated areas and negative correlations with occupied urban areas with some vegetation. Land surface temperature correlated negatively with VL incidence in both periods. Object-oriented classification can be useful to characterize landscape features associated with VL in urban areas and to help identify risk areas in order to prioritize interventions.

Classification of Parkinson's disease utilizing multi-edit nearest-neighbor and ensemble learning algorithms with speech samples.

PubMed

Zhang, He-Hua; Yang, Liuyang; Liu, Yuchuan; Wang, Pin; Yin, Jun; Li, Yongming; Qiu, Mingguo; Zhu, Xueru; Yan, Fang

2016-11-16

The use of speech based data in the classification of Parkinson disease (PD) has been shown to provide an effect, non-invasive mode of classification in recent years. Thus, there has been an increased interest in speech pattern analysis methods applicable to Parkinsonism for building predictive tele-diagnosis and tele-monitoring models. One of the obstacles in optimizing classifications is to reduce noise within the collected speech samples, thus ensuring better classification accuracy and stability. While the currently used methods are effect, the ability to invoke instance selection has been seldomly examined. In this study, a PD classification algorithm was proposed and examined that combines a multi-edit-nearest-neighbor (MENN) algorithm and an ensemble learning algorithm. First, the MENN algorithm is applied for selecting optimal training speech samples iteratively, thereby obtaining samples with high separability. Next, an ensemble learning algorithm, random forest (RF) or decorrelated neural network ensembles (DNNE), is used to generate trained samples from the collected training samples. Lastly, the trained ensemble learning algorithms are applied to the test samples for PD classification. This proposed method was examined using a more recently deposited public datasets and compared against other currently used algorithms for validation. Experimental results showed that the proposed algorithm obtained the highest degree of improved classification accuracy (29.44%) compared with the other algorithm that was examined. Furthermore, the MENN algorithm alone was found to improve classification accuracy by as much as 45.72%. Moreover, the proposed algorithm was found to exhibit a higher stability, particularly when combining the MENN and RF algorithms. This study showed that the proposed method could improve PD classification when using speech data and can be applied to future studies seeking to improve PD classification methods.

An Optimization-based Framework to Learn Conditional Random Fields for Multi-label Classification

PubMed Central

Naeini, Mahdi Pakdaman; Batal, Iyad; Liu, Zitao; Hong, CharmGil; Hauskrecht, Milos

2015-01-01

This paper studies multi-label classification problem in which data instances are associated with multiple, possibly high-dimensional, label vectors. This problem is especially challenging when labels are dependent and one cannot decompose the problem into a set of independent classification problems. To address the problem and properly represent label dependencies we propose and study a pairwise conditional random Field (CRF) model. We develop a new approach for learning the structure and parameters of the CRF from data. The approach maximizes the pseudo likelihood of observed labels and relies on the fast proximal gradient descend for learning the structure and limited memory BFGS for learning the parameters of the model. Empirical results on several datasets show that our approach outperforms several multi-label classification baselines, including recently published state-of-the-art methods. PMID:25927015

Advances in Patient Classification for Traditional Chinese Medicine: A Machine Learning Perspective

PubMed Central

Zhao, Changbo; Li, Guo-Zheng; Wang, Chengjun; Niu, Jinling

2015-01-01

As a complementary and alternative medicine in medical field, traditional Chinese medicine (TCM) has drawn great attention in the domestic field and overseas. In practice, TCM provides a quite distinct methodology to patient diagnosis and treatment compared to western medicine (WM). Syndrome (ZHENG or pattern) is differentiated by a set of symptoms and signs examined from an individual by four main diagnostic methods: inspection, auscultation and olfaction, interrogation, and palpation which reflects the pathological and physiological changes of disease occurrence and development. Patient classification is to divide patients into several classes based on different criteria. In this paper, from the machine learning perspective, a survey on patient classification issue will be summarized on three major aspects of TCM: sign classification, syndrome differentiation, and disease classification. With the consideration of different diagnostic data analyzed by different computational methods, we present the overview for four subfields of TCM diagnosis, respectively. For each subfield, we design a rectangular reference list with applications in the horizontal direction and machine learning algorithms in the longitudinal direction. According to the current development of objective TCM diagnosis for patient classification, a discussion of the research issues around machine learning techniques with applications to TCM diagnosis is given to facilitate the further research for TCM patient classification. PMID:26246834

Cooperative Learning for Distributed In-Network Traffic Classification

NASA Astrophysics Data System (ADS)

Joseph, S. B.; Loo, H. R.; Ismail, I.; Andromeda, T.; Marsono, M. N.

2017-04-01

Inspired by the concept of autonomic distributed/decentralized network management schemes, we consider the issue of information exchange among distributed network nodes to network performance and promote scalability for in-network monitoring. In this paper, we propose a cooperative learning algorithm for propagation and synchronization of network information among autonomic distributed network nodes for online traffic classification. The results show that network nodes with sharing capability perform better with a higher average accuracy of 89.21% (sharing data) and 88.37% (sharing clusters) compared to 88.06% for nodes without cooperative learning capability. The overall performance indicates that cooperative learning is promising for distributed in-network traffic classification.

A Study on the Learning Processes in Discrimination Shift Learning of Children with Mental Retardation: From the Point of Developmental View of "Logical Manipulation by Classification."

ERIC Educational Resources Information Center

Kanno, Atsushi

1989-01-01

The study was designed to investigate the learning processes in discrimination shift learning, in terms of developmental views of "logical manipulation by classification." Tasks comparing sizes of intradimensional value-classes and comparing sizes of interdimensional value-classes were devised in order to measure subjects' levels of…

An alternative to soil taxonomy for describing key soil characteristics

USGS Publications Warehouse

Duniway, Michael C.; Miller, Mark E.; Brown, Joel R.; Toevs, Gordon

2013-01-01

is not a simple task. Furthermore, because the US system of soil taxonomy is not applied universally, its utility as a means for effectively describing soil characteristics to readers in other countries is limited. Finally, and most importantly, even at the finest level of soil classification there are often large within-taxa variations in critical properties that can determine ecosystem responses to drivers such as climate and land-use change.

A Brief History of Soil Mapping and Classification in the USA

NASA Astrophysics Data System (ADS)

Brevik, Eric C.; Hartemink, Alfred E.

2014-05-01

Soil maps show the distribution of soils across an area but also depict soil science theory and ideas on soil formation and classification at the time the maps were created. The national soil mapping program in the USA was established in 1899. The first nation-wide soil map was published by M. Whitney in 1909 and showed soil provinces that were largely based on geology. In 1912, G.N. Coffey published the first country-wide map based on soil properties. The map showed 5 broad soil units that used parent material, color and drainage as diagnostic criteria. The 1913 national map was produced by C.F. Marbut, H.H. Bennett, J.E. Lapham, and M.H. Lapham and showed broad physiographic units that were further subdivided into soil series, soil classes and soil types. In 1935, Marbut drafted a series of maps based on soil properties, but these maps were replaced as official U.S. soil maps in 1938 with the work of M. Baldwin, C.E. Kellogg, and J. Thorp. A series of soil maps similar to modern USA maps appeared in the 1960s with the 7th Approximation followed by revisions with the 1975 and 1999 editions of Soil Taxonomy. This review has shown that soil maps in the United States produced since the early 1900s moved initially from a geologic-based concept to a pedologic concept of soils. Later changes were from property-based systems to process-based, and then back to property-based. The information in this presentation is based on Brevik and Hartemink (2013). Brevik, E.C., and A.E. Hartemink. 2013. Soil Maps of the United States of America. Soil Science Society of America Journal 77:1117-1132. doi:10.2136/sssaj2012.0390.

Automated diagnosis of myositis from muscle ultrasound: Exploring the use of machine learning and deep learning methods

PubMed Central

Burlina, Philippe; Billings, Seth; Joshi, Neil

2017-01-01

Objective To evaluate the use of ultrasound coupled with machine learning (ML) and deep learning (DL) techniques for automated or semi-automated classification of myositis. Methods Eighty subjects comprised of 19 with inclusion body myositis (IBM), 14 with polymyositis (PM), 14 with dermatomyositis (DM), and 33 normal (N) subjects were included in this study, where 3214 muscle ultrasound images of 7 muscles (observed bilaterally) were acquired. We considered three problems of classification including (A) normal vs. affected (DM, PM, IBM); (B) normal vs. IBM patients; and (C) IBM vs. other types of myositis (DM or PM). We studied the use of an automated DL method using deep convolutional neural networks (DL-DCNNs) for diagnostic classification and compared it with a semi-automated conventional ML method based on random forests (ML-RF) and “engineered” features. We used the known clinical diagnosis as the gold standard for evaluating performance of muscle classification. Results The performance of the DL-DCNN method resulted in accuracies ± standard deviation of 76.2% ± 3.1% for problem (A), 86.6% ± 2.4% for (B) and 74.8% ± 3.9% for (C), while the ML-RF method led to accuracies of 72.3% ± 3.3% for problem (A), 84.3% ± 2.3% for (B) and 68.9% ± 2.5% for (C). Conclusions This study demonstrates the application of machine learning methods for automatically or semi-automatically classifying inflammatory muscle disease using muscle ultrasound. Compared to the conventional random forest machine learning method used here, which has the drawback of requiring manual delineation of muscle/fat boundaries, DCNN-based classification by and large improved the accuracies in all classification problems while providing a fully automated approach to classification. PMID:28854220

Automated diagnosis of myositis from muscle ultrasound: Exploring the use of machine learning and deep learning methods.

PubMed

Burlina, Philippe; Billings, Seth; Joshi, Neil; Albayda, Jemima

2017-01-01

To evaluate the use of ultrasound coupled with machine learning (ML) and deep learning (DL) techniques for automated or semi-automated classification of myositis. Eighty subjects comprised of 19 with inclusion body myositis (IBM), 14 with polymyositis (PM), 14 with dermatomyositis (DM), and 33 normal (N) subjects were included in this study, where 3214 muscle ultrasound images of 7 muscles (observed bilaterally) were acquired. We considered three problems of classification including (A) normal vs. affected (DM, PM, IBM); (B) normal vs. IBM patients; and (C) IBM vs. other types of myositis (DM or PM). We studied the use of an automated DL method using deep convolutional neural networks (DL-DCNNs) for diagnostic classification and compared it with a semi-automated conventional ML method based on random forests (ML-RF) and "engineered" features. We used the known clinical diagnosis as the gold standard for evaluating performance of muscle classification. The performance of the DL-DCNN method resulted in accuracies ± standard deviation of 76.2% ± 3.1% for problem (A), 86.6% ± 2.4% for (B) and 74.8% ± 3.9% for (C), while the ML-RF method led to accuracies of 72.3% ± 3.3% for problem (A), 84.3% ± 2.3% for (B) and 68.9% ± 2.5% for (C). This study demonstrates the application of machine learning methods for automatically or semi-automatically classifying inflammatory muscle disease using muscle ultrasound. Compared to the conventional random forest machine learning method used here, which has the drawback of requiring manual delineation of muscle/fat boundaries, DCNN-based classification by and large improved the accuracies in all classification problems while providing a fully automated approach to classification.

A Classification of Remote Sensing Image Based on Improved Compound Kernels of Svm

NASA Astrophysics Data System (ADS)

Zhao, Jianing; Gao, Wanlin; Liu, Zili; Mou, Guifen; Lu, Lin; Yu, Lina

The accuracy of RS classification based on SVM which is developed from statistical learning theory is high under small number of train samples, which results in satisfaction of classification on RS using SVM methods. The traditional RS classification method combines visual interpretation with computer classification. The accuracy of the RS classification, however, is improved a lot based on SVM method, because it saves much labor and time which is used to interpret images and collect training samples. Kernel functions play an important part in the SVM algorithm. It uses improved compound kernel function and therefore has a higher accuracy of classification on RS images. Moreover, compound kernel improves the generalization and learning ability of the kernel.

[Review on water eco-environment in vegetation restoration in Loess Plateau].

PubMed

Hu, Liangjun; Shao, Mingan

2002-08-01

Water is the crucial factor influencing vegetation restoration and eco-environmental reconstruction in Loess Plateau region. In this paper, the previous studies on water eco-environment under vegetation construction were summarized from seven aspects, i.e., soil water resource, background of soil water, dynamics of soil water, dry soil layer, relationship between soil water and vegetarian productivity, classification of soil water position, and strategy for vegetation construction. Meanwhile, some problems in the relevant researches were pointed out and discussed.

The Influence of Processing Soil With a Coffee Grinder on Soil Classification

DTIC Science & Technology

2015-01-20

shaker, sieves , coffee grinder, plastic limit tool, bowls, spatulas, and scoops. To classify soils, a dry sieve analysis is performed, as is a Plastic...processed with the coffee grinder for 90 seconds as described above. Sieve analysis using the wet preparation method was used to test and classify the soils...one 90 second cycle of Elevator Soil Figure 3: The blades after three 90 second cycles of Elevator Soil 71Page 4.2 Ottawa Sand Dry Sieve Analysis

Classification of the Regional Ionospheric Disturbance Based on Machine Learning Techniques

NASA Astrophysics Data System (ADS)

Terzi, Merve Begum; Arikan, Orhan; Karatay, Secil; Arikan, Feza; Gulyaeva, Tamara

2016-08-01

In this study, Total Electron Content (TEC) estimated from GPS receivers is used to model the regional and local variability that differs from global activity along with solar and geomagnetic indices. For the automated classification of regional disturbances, a classification technique based on a robust machine learning technique that have found wide spread use, Support Vector Machine (SVM) is proposed. Performance of developed classification technique is demonstrated for midlatitude ionosphere over Anatolia using TEC estimates generated from GPS data provided by Turkish National Permanent GPS Network (TNPGN-Active) for solar maximum year of 2011. As a result of implementing developed classification technique to Global Ionospheric Map (GIM) TEC data, which is provided by the NASA Jet Propulsion Laboratory (JPL), it is shown that SVM can be a suitable learning method to detect anomalies in TEC variations.

Large-Scale Machine Learning for Classification and Search

ERIC Educational Resources Information Center

Liu, Wei

2012-01-01

With the rapid development of the Internet, nowadays tremendous amounts of data including images and videos, up to millions or billions, can be collected for training machine learning models. Inspired by this trend, this thesis is dedicated to developing large-scale machine learning techniques for the purpose of making classification and nearest…

Event-Related fMRI of Category Learning: Differences in Classification and Feedback Networks

ERIC Educational Resources Information Center

Little, Deborah M.; Shin, Silvia S.; Sisco, Shannon M.; Thulborn, Keith R.

2006-01-01

Eighteen healthy young adults underwent event-related (ER) functional magnetic resonance imaging (fMRI) of the brain while performing a visual category learning task. The specific category learning task required subjects to extract the rules that guide classification of quasi-random patterns of dots into categories. Following each classification…

Test-Enhanced Learning of Natural Concepts: Effects on Recognition Memory, Classification, and Metacognition

ERIC Educational Resources Information Center

Jacoby, Larry L.; Wahlheim, Christopher N.; Coane, Jennifer H.

2010-01-01

Three experiments examined testing effects on learning of natural concepts and metacognitive assessments of such learning. Results revealed that testing enhanced recognition memory and classification accuracy for studied and novel exemplars of bird families on immediate and delayed tests. These effects depended on the balance of study and test…

Subsurface event detection and classification using Wireless Signal Networks.

PubMed

Yoon, Suk-Un; Ghazanfari, Ehsan; Cheng, Liang; Pamukcu, Sibel; Suleiman, Muhannad T

2012-11-05

Subsurface environment sensing and monitoring applications such as detection of water intrusion or a landslide, which could significantly change the physical properties of the host soil, can be accomplished using a novel concept, Wireless Signal Networks (WSiNs). The wireless signal networks take advantage of the variations of radio signal strength on the distributed underground sensor nodes of WSiNs to monitor and characterize the sensed area. To characterize subsurface environments for event detection and classification, this paper provides a detailed list and experimental data of soil properties on how radio propagation is affected by soil properties in subsurface communication environments. Experiments demonstrated that calibrated wireless signal strength variations can be used as indicators to sense changes in the subsurface environment. The concept of WSiNs for the subsurface event detection is evaluated with applications such as detection of water intrusion, relative density change, and relative motion using actual underground sensor nodes. To classify geo-events using the measured signal strength as a main indicator of geo-events, we propose a window-based minimum distance classifier based on Bayesian decision theory. The window-based classifier for wireless signal networks has two steps: event detection and event classification. With the event detection, the window-based classifier classifies geo-events on the event occurring regions that are called a classification window. The proposed window-based classification method is evaluated with a water leakage experiment in which the data has been measured in laboratory experiments. In these experiments, the proposed detection and classification method based on wireless signal network can detect and classify subsurface events.

Subsurface Event Detection and Classification Using Wireless Signal Networks

PubMed Central

Yoon, Suk-Un; Ghazanfari, Ehsan; Cheng, Liang; Pamukcu, Sibel; Suleiman, Muhannad T.

2012-01-01

Subsurface environment sensing and monitoring applications such as detection of water intrusion or a landslide, which could significantly change the physical properties of the host soil, can be accomplished using a novel concept, Wireless Signal Networks (WSiNs). The wireless signal networks take advantage of the variations of radio signal strength on the distributed underground sensor nodes of WSiNs to monitor and characterize the sensed area. To characterize subsurface environments for event detection and classification, this paper provides a detailed list and experimental data of soil properties on how radio propagation is affected by soil properties in subsurface communication environments. Experiments demonstrated that calibrated wireless signal strength variations can be used as indicators to sense changes in the subsurface environment. The concept of WSiNs for the subsurface event detection is evaluated with applications such as detection of water intrusion, relative density change, and relative motion using actual underground sensor nodes. To classify geo-events using the measured signal strength as a main indicator of geo-events, we propose a window-based minimum distance classifier based on Bayesian decision theory. The window-based classifier for wireless signal networks has two steps: event detection and event classification. With the event detection, the window-based classifier classifies geo-events on the event occurring regions that are called a classification window. The proposed window-based classification method is evaluated with a water leakage experiment in which the data has been measured in laboratory experiments. In these experiments, the proposed detection and classification method based on wireless signal network can detect and classify subsurface events. PMID:23202191

Land cover classification in multispectral imagery using clustering of sparse approximations over learned feature dictionaries

DOE PAGES

Moody, Daniela I.; Brumby, Steven P.; Rowland, Joel C.; ...

2014-12-09

We present results from an ongoing effort to extend neuromimetic machine vision algorithms to multispectral data using adaptive signal processing combined with compressive sensing and machine learning techniques. Our goal is to develop a robust classification methodology that will allow for automated discretization of the landscape into distinct units based on attributes such as vegetation, surface hydrological properties, and topographic/geomorphic characteristics. We use a Hebbian learning rule to build spectral-textural dictionaries that are tailored for classification. We learn our dictionaries from millions of overlapping multispectral image patches and then use a pursuit search to generate classification features. Land cover labelsmore » are automatically generated using unsupervised clustering of sparse approximations (CoSA). We demonstrate our method on multispectral WorldView-2 data from a coastal plain ecosystem in Barrow, Alaska. We explore learning from both raw multispectral imagery and normalized band difference indices. We explore a quantitative metric to evaluate the spectral properties of the clusters in order to potentially aid in assigning land cover categories to the cluster labels. In this study, our results suggest CoSA is a promising approach to unsupervised land cover classification in high-resolution satellite imagery.« less

Learning accurate very fast decision trees from uncertain data streams

NASA Astrophysics Data System (ADS)

Liang, Chunquan; Zhang, Yang; Shi, Peng; Hu, Zhengguo

2015-12-01

Most existing works on data stream classification assume the streaming data is precise and definite. Such assumption, however, does not always hold in practice, since data uncertainty is ubiquitous in data stream applications due to imprecise measurement, missing values, privacy protection, etc. The goal of this paper is to learn accurate decision tree models from uncertain data streams for classification analysis. On the basis of very fast decision tree (VFDT) algorithms, we proposed an algorithm for constructing an uncertain VFDT tree with classifiers at tree leaves (uVFDTc). The uVFDTc algorithm can exploit uncertain information effectively and efficiently in both the learning and the classification phases. In the learning phase, it uses Hoeffding bound theory to learn from uncertain data streams and yield fast and reasonable decision trees. In the classification phase, at tree leaves it uses uncertain naive Bayes (UNB) classifiers to improve the classification performance. Experimental results on both synthetic and real-life datasets demonstrate the strong ability of uVFDTc to classify uncertain data streams. The use of UNB at tree leaves has improved the performance of uVFDTc, especially the any-time property, the benefit of exploiting uncertain information, and the robustness against uncertainty.

Group-Based Active Learning of Classification Models.

PubMed

Luo, Zhipeng; Hauskrecht, Milos

2017-05-01

Learning of classification models from real-world data often requires additional human expert effort to annotate the data. However, this process can be rather costly and finding ways of reducing the human annotation effort is critical for this task. The objective of this paper is to develop and study new ways of providing human feedback for efficient learning of classification models by labeling groups of examples. Briefly, unlike traditional active learning methods that seek feedback on individual examples, we develop a new group-based active learning framework that solicits label information on groups of multiple examples. In order to describe groups in a user-friendly way, conjunctive patterns are used to compactly represent groups. Our empirical study on 12 UCI data sets demonstrates the advantages and superiority of our approach over both classic instance-based active learning work, as well as existing group-based active-learning methods.

Framing a future for soil science education.

NASA Astrophysics Data System (ADS)

Field, Damien

2017-04-01

The emerging concept of Global Soil Security highlights the need to have a renewed education framework that addresses the needs of those who want to; 1) know soil, 2) know of soil, and/or 3) be aware of soil. Those who know soil are soil science discipline experts and are concerned with soil as an object of study. With their discipline expertise focusing on what soil's are capable of they would be brokers of soil knowledge to those who know of soil. The connection with soil by the those in the second group focuses on the soil's utility and are responsible for managing the functionality and condition of the soil, the obvious example are farmers and agronomists. Reconnecting society with soil illustrates those who are members of the third group, i.e. those who are aware of soil. This is predicated on concepts of 'care' and is founded in the notion of beauty and utility. The utility is concerned with soil providing good Quality, clean food, or a source of pharmaceuticals. Soil also provides a place for recreation and those aware of soil know who this contributes to human health. The teaching-research-industry-learning (TRIL) nexus has been used to develop a framework for the learning and teaching of soil science applicable to a range of recipients, particularly campus-based students and practicing farm advisors. Consultation with academics, industry and professionals, by means of online (Delphi Study) and face-to-face forums, developed a heavily content-rich core body of knowledge (CBoK) relevant to industry, satisfying those who; know, and know of soil. Integrating the multidisciplinary approach in soil science teaching is a future aspiration, and will enable the development of curriculum that incorporates those who 'care' for soil. In the interim the application of the TRIL model allows the development of a learning framework more suited to real word needs. The development of a learning framework able to meet industry needs includes authentic complex scenarios that will also benefit student learning.

An Active Learning Framework for Hyperspectral Image Classification Using Hierarchical Segmentation

NASA Technical Reports Server (NTRS)

Zhang, Zhou; Pasolli, Edoardo; Crawford, Melba M.; Tilton, James C.

2015-01-01

Augmenting spectral data with spatial information for image classification has recently gained significant attention, as classification accuracy can often be improved by extracting spatial information from neighboring pixels. In this paper, we propose a new framework in which active learning (AL) and hierarchical segmentation (HSeg) are combined for spectral-spatial classification of hyperspectral images. The spatial information is extracted from a best segmentation obtained by pruning the HSeg tree using a new supervised strategy. The best segmentation is updated at each iteration of the AL process, thus taking advantage of informative labeled samples provided by the user. The proposed strategy incorporates spatial information in two ways: 1) concatenating the extracted spatial features and the original spectral features into a stacked vector and 2) extending the training set using a self-learning-based semi-supervised learning (SSL) approach. Finally, the two strategies are combined within an AL framework. The proposed framework is validated with two benchmark hyperspectral datasets. Higher classification accuracies are obtained by the proposed framework with respect to five other state-of-the-art spectral-spatial classification approaches. Moreover, the effectiveness of the proposed pruning strategy is also demonstrated relative to the approaches based on a fixed segmentation.

Classification and evaluation for forest sites on the Eastern Highland Rim and Pennyroyal.

Treesearch

Glendon W. Smalley

1983-01-01

Presents a comprehensive forest site classification system for the Eastern Highland Rim and Pennyroyal in north Alabama, east-central Tennessee, and central Kentucky. The system is based on physiography, geology, soils, topography, and vegetation.

Teaching with Moodle in Soil Science

NASA Astrophysics Data System (ADS)

Roca, Núria

2014-05-01

Soil is a 3-dimensional body with properties that reflect the impact of climate, vegetation, fauna, man and topography on the soil's parent material over a variable time span. Therefore, soil is integral to many ecological and social systems and it holds potential solutions for many of the world's economic and scientific problems as climate change or scarcity of food and water. The teaching of Soil Science, as a natural science in its own right, requires principles that reflect the unique features and behaviour of soil and the practices of soil scientists. It could be argued that a unique set of teaching practices applies to Soil Science; however specific teaching practices are scarce in literature. The present work was triggered by the need to develop new techniques of teaching to speed up the learning process and to experiment with new methods of teaching. For such, it is necessary to adopt virtual learning environment to new learning requirements regarding Soil Science. This paper proposes a set of e-teaching techniques (as questionnaires, chats as well as forums) introduced in Moodle virtual learning Environment in order to increase student motivation and interest in Soil Science. Such technologies can be used to: a)Increase the amount of time a teacher allots for student reflection after asking a question and before a student responds (wait-time). This practice increases the quantity and quality of students' answers. The students give longer responses, students give more evidence for their ideas and conclusions, students speculate and hypothesize more and more students participated in responding. Furthermore, students ask more questions and talk more to other students. b)Improve active learning, an essential paradigm in education. In contrast to learning-before-doing, we propose to focus on learning-in-doing, a model where learners are increasingly involved in the authentic practices of communities through learning conversations and activities involving expert practitioners, educators and peers. c)Introduce the specific specialised technical language (jargon) gradually. The excessive use of Soil Science jargon confuses students and frequently put obstacles in the way of learning. d)Encourage the students to take responsibility for their learning, continuous assessment with direct error correction and content feedback and peer review with comments sent to forum. The student interest to learn using e-project is clearly strong.

Traditional knowledge among Zapotecs of Sierra Madre Del Sur, Oaxaca. Does it represent a base for plant resources management and conservation?

PubMed Central

2012-01-01

Traditional classification systems represent cognitive processes of human cultures in the world. It synthesizes specific conceptions of nature, as well as cumulative learning, beliefs and customs that are part of a particular human community or society. Traditional knowledge has been analyzed from different viewpoints, one of which corresponds to the analysis of ethnoclassifications. In this work, a brief analysis of the botanical traditional knowledge among Zapotecs of the municipality of San Agustin Loxicha, Oaxaca was conducted. The purposes of this study were: a) to analyze the traditional ecological knowledge of local plant resources through the folk classification of both landscapes and plants and b) to determine the role that this knowledge has played in plant resource management and conservation. The study was developed in five communities of San Agustín Loxicha. From field trips, plant specimens were collected and showed to local people in order to get the Spanish or Zapotec names; through interviews with local people, we obtained names and identified classification categories of plants, vegetation units, and soil types. We found a logic structure in Zapotec plant names, based on linguistic terms, as well as morphological and ecological caracteristics. We followed the classification principles proposed by Berlin [6] in order to build a hierarchical structure of life forms, names and other characteristics mentioned by people. We recorded 757 plant names. Most of them (67%) have an equivalent Zapotec name and the remaining 33% had mixed names with Zapotec and Spanish terms. Plants were categorized as native plants, plants introduced in pre-Hispanic times, or plants introduced later. All of them are grouped in a hierarchical classification, which include life form, generic, specific, and varietal categories. Monotypic and polytypic names are used to further classify plants. This holistic classification system plays an important role for local people in many aspects: it helps to organize and make sense of the diversity, to understand the interrelation among plants–soil–vegetation and to classify their physical space since they relate plants with a particular vegetation unit and a kind of soil. The locals also make a rational use of these elements, because they know which crops can grow in any vegetation unit, or which places are indicated to recollect plants. These aspects are interconnected and could be fundamental for a rational use and management of plant resources. PMID:22789155

New wideband radar target classification method based on neural learning and modified Euclidean metric

NASA Astrophysics Data System (ADS)

Jiang, Yicheng; Cheng, Ping; Ou, Yangkui

2001-09-01

A new method for target classification of high-range resolution radar is proposed. It tries to use neural learning to obtain invariant subclass features of training range profiles. A modified Euclidean metric based on the Box-Cox transformation technique is investigated for Nearest Neighbor target classification improvement. The classification experiments using real radar data of three different aircraft have demonstrated that classification error can reduce 8% if this method proposed in this paper is chosen instead of the conventional method. The results of this paper have shown that by choosing an optimized metric, it is indeed possible to reduce the classification error without increasing the number of samples.

Variability of the soil-to-plant radiocaesium transfer factor for Japanese soils predicted with soil and plant properties.

PubMed

Uematsu, Shinichiro; Vandenhove, Hildegarde; Sweeck, Lieve; Van Hees, May; Wannijn, Jean; Smolders, Erik

2016-03-01

Food chain contamination with radiocaesium (RCs) in the aftermath of the Fukushima accident calls for an analysis of the specific factors that control the RCs transfer. Here, soil-to-plant transfer factors (TF) of RCs for grass were predicted from the potassium concentration in soil solution (mK) and the Radiocaesium Interception Potential (RIP) of the soil using existing mechanistic models. The mK and RIP were (a) either measured for 37 topsoils collected from the Fukushima accident affected area or (b) predicted from the soil clay content and the soil exchangeable potassium content using the models that had been calibrated for European soils. An average ammonium concentration was used throughout in the prediction. The measured RIP ranged 14-fold and measured mK varied 37-fold among the soils. The measured RIP was lower than the RIP predicted from the soil clay content likely due to the lower content of weathered micas in the clay fraction of Japanese soils. Also the measured mK was lower than that predicted. As a result, the predicted TFs relying on the measured RIP and mK were, on average, about 22-fold larger than the TFs predicted using the European calibrated models. The geometric mean of the measured TFs for grass in the affected area (N = 82) was in the middle of both. The TFs were poorly related to soil classification classes, likely because soil fertility (mK) was obscuring the effects of the soil classification related to the soil mineralogy (RIP). This study suggests that, on average, Japanese soils are more vulnerable than European soils at equal soil clay and exchangeable K content. The affected regions will be targeted for refined model validation. Copyright © 2015 Elsevier Ltd. All rights reserved.

Review and Future Research Directions about Major Monitoring Method of Soil Erosion

NASA Astrophysics Data System (ADS)

LI, Yue; Bai, Xiaoyong; Tian, Yichao; Luo, Guangjie

2017-05-01

Soil erosion is a highly serious ecological problem that occurs worldwide. Hence,scientific methods for accurate monitoring are needed to obtain soil erosion data. At present,numerous methods on soil erosion monitoring are being used internationally. In this paper, wepresent a systematic classification of these methods based on the date of establishment andtype of approach. This classification comprises five categories: runoff plot method, erosion pinmethod, radionuclide tracer method, model estimation, and 3S technology combined method.The backgrounds of their establishment are briefly introduced, the history of their developmentis reviewed, and the conditions for their application are enumerated. Their respectiveadvantages and disadvantages are compared and analysed, and future prospects regarding theirdevelopment are discussed. We conclude that the methods of soil erosion monitoring in the past 100 years of their development constantly considered the needs of the time. According to the progress of soil erosion monitoring technology throughout its history, we predict that the future trend in this field would move toward the development of quantitative, precise, and composite methods. This report serves as a valuable reference for scientific and technological workers globally, especially those engaged in soil erosion research.

Global Optimization Ensemble Model for Classification Methods

PubMed Central

Anwar, Hina; Qamar, Usman; Muzaffar Qureshi, Abdul Wahab

2014-01-01

Supervised learning is the process of data mining for deducing rules from training datasets. A broad array of supervised learning algorithms exists, every one of them with its own advantages and drawbacks. There are some basic issues that affect the accuracy of classifier while solving a supervised learning problem, like bias-variance tradeoff, dimensionality of input space, and noise in the input data space. All these problems affect the accuracy of classifier and are the reason that there is no global optimal method for classification. There is not any generalized improvement method that can increase the accuracy of any classifier while addressing all the problems stated above. This paper proposes a global optimization ensemble model for classification methods (GMC) that can improve the overall accuracy for supervised learning problems. The experimental results on various public datasets showed that the proposed model improved the accuracy of the classification models from 1% to 30% depending upon the algorithm complexity. PMID:24883382

Spectral signature selection for mapping unvegetated soils

NASA Technical Reports Server (NTRS)

May, G. A.; Petersen, G. W.

1975-01-01

Airborne multispectral scanner data covering the wavelength interval from 0.40-2.60 microns were collected at an altitude of 1000 m above the terrain in southeastern Pennsylvania. Uniform training areas were selected within three sites from this flightline. Soil samples were collected from each site and a procedure developed to allow assignment of scan line and element number from the multispectral scanner data to each sampling location. These soil samples were analyzed on a spectrophotometer and laboratory spectral signatures were derived. After correcting for solar radiation and atmospheric attenuation, the laboratory signatures were compared to the spectral signatures derived from these same soils using multispectral scanner data. Both signatures were used in supervised and unsupervised classification routines. Computer-generated maps using the laboratory and multispectral scanner derived signatures resulted in maps that were similar to maps resulting from field surveys. Approximately 90% agreement was obtained between classification maps produced using multispectral scanner derived signatures and laboratory derived signatures.

Influence of Long-term Application of Feedlot Manure Amendments on Water Repellency of a Clay Loam Soil.

PubMed

Miller, Jim J; Beasley, Bruce W; Hazendonk, Paul; Drury, Craig F; Chanasyk, David S

2017-05-01

Long-term application of feedlot manure to cropland may increase the quantity of soil organic carbon (C) and change its quality, which may influence soil water repellency. The objective was to determine the influence of feedlot manure type (stockpiled vs. composted), bedding material (straw [ST] vs. woodchips [WD]), and application rate (13, 39, or 77 Mg ha) on repellency of a clay loam soil after 17 annual applications. The repellency was determined on all 14 treatments using the water repellency index ( index), the water drop penetration time (WDPT) method, and molarity of ethanol (MED) test. The C composition of particulate organic matter in soil of five selected treatments after 16 annual applications was also determined using C nuclear magnetic resonance-direct polarization with magic-angle spinning (NMR-DPMAS). Manure type had no significant ( > 0.05) effect on index and WDPT, and MED classification was similar. Mean index and WDPT values were significantly greater and MED classification more hydrophobic for WD than ST. Application rate had no effect on the index, but WDPT was significantly greater and MED classification more hydrophobic with increasing application rate. Strong ( > 0.7) but nonsignificant positive correlations were found between index and WDPT versus hydrophobic (alkyl + aromatic) C, lignin at 74 ppm (O-alkyl), and unspecified aromatic compounds at 144 ppm. Specific aromatic compounds also contributed more to repellency than alkyl, O-alkyl, and carbonyl compounds. Overall, all three methods consistently showed that repellency was greater for WD- than ST-amended clay loam soil, but manure type had no effect. Copyright © by the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America, Inc.

STIR Proposal For Research Area 2.1.2 Surface Energy Balance: Transient Soil Density Impacts Land Surface Characteristics and Characterization

DTIC Science & Technology

2015-12-22

not shown). The relatively small differences were likely associated with differences in surface albedo and longwave radiation from soil surface. Ground...SECURITY CLASSIFICATION OF: Soil density is commonly treated as static in studies on land surface property dynamics. Magnitudes of errors associated...with this assumption are largely unknown. Objectives of this preliminary investigation were to: i) quantify effects of soil density variation on soil

Chlorophyll fluorescence as a tool for nutrient status identification in rapeseed plants.

PubMed

Kalaji, Hazem M; Bąba, Wojciech; Gediga, Krzysztof; Goltsev, Vasilij; Samborska, Izabela A; Cetner, Magdalena D; Dimitrova, Stella; Piszcz, Urszula; Bielecki, Krzysztof; Karmowska, Kamila; Dankov, Kolyo; Kompała-Bąba, Agnieszka

2018-06-01

In natural conditions, plants growth and development depends on environmental conditions, including the availability of micro- and macroelements in the soil. Nutrient status should thus be examined not by establishing the effects of single nutrient deficiencies on the physiological state of the plant but by combinations of them. Differences in the nutrient content significantly affect the photochemical process of photosynthesis therefore playing a crucial role in plants growth and development. In this work, an attempt was made to find a connection between element content in (i) different soils, (ii) plant leaves, grown on these soils and (iii) changes in selected chlorophyll a fluorescence parameters, in order to find a method for early detection of plant stress resulting from the combination of nutrient status in natural conditions. To achieve this goal, a mathematical procedure was used which combines principal component analysis (a tool for the reduction of data complexity), hierarchical k-means (a classification method) and a machine-learning method-super-organising maps. Differences in the mineral content of soil and plant leaves resulted in functional changes in the photosynthetic machinery that can be measured by chlorophyll a fluorescent signals. Five groups of patterns in the chlorophyll fluorescent parameters were established: the 'no deficiency', Fe-specific deficiency, slight, moderate and strong deficiency. Unfavourable development in groups with nutrient deficiency of any kind was reflected by a strong increase in F o and ΔV/Δt 0 and decline in φ Po , φ Eo δ Ro and φ Ro . The strong deficiency group showed the suboptimal development of the photosynthetic machinery, which affects both PSII and PSI. The nutrient-deficient groups also differed in antenna complex organisation. Thus, our work suggests that the chlorophyll fluorescent method combined with machine-learning methods can be highly informative and in some cases, it can replace much more expensive and time-consuming procedures such as chemometric analyses.

Combining Unsupervised and Supervised Classification to Build User Models for Exploratory Learning Environments

ERIC Educational Resources Information Center

Amershi, Saleema; Conati, Cristina

2009-01-01

In this paper, we present a data-based user modeling framework that uses both unsupervised and supervised classification to build student models for exploratory learning environments. We apply the framework to build student models for two different learning environments and using two different data sources (logged interface and eye-tracking data).…

2014 Bio-Acoustics Data Challenge for the International Community on Machine Learning and Bioacoustics

DTIC Science & Technology

2014-09-30

This ONR grant promotes the development and application of advanced machine learning techniques for detection and classification of marine mammal...sounds. The objective is to engage a broad community of data scientists in the development and application of advanced machine learning techniques for detection and classification of marine mammal sounds.

Learning and retention through predictive inference and classification.

PubMed

Sakamoto, Yasuaki; Love, Bradley C

2010-12-01

Work in category learning addresses how humans acquire knowledge and, thus, should inform classroom practices. In two experiments, we apply and evaluate intuitions garnered from laboratory-based research in category learning to learning tasks situated in an educational context. In Experiment 1, learning through predictive inference and classification were compared for fifth-grade students using class-related materials. Making inferences about properties of category members and receiving feedback led to the acquisition of both queried (i.e., tested) properties and nonqueried properties that were correlated with a queried property (e.g., even if not queried, students learned about a species' habitat because it correlated with a queried property, like the species' size). In contrast, classifying items according to their species and receiving feedback led to knowledge of only the property most diagnostic of category membership. After multiple-day delay, the fifth-graders who learned through inference selectively retained information about the queried properties, and the fifth-graders who learned through classification retained information about the diagnostic property, indicating a role for explicit evaluation in establishing memories. Overall, inference learning resulted in fewer errors, better retention, and more liking of the categories than did classification learning. Experiment 2 revealed that querying a property only a few times was enough to manifest the full benefits of inference learning in undergraduate students. These results suggest that classroom teaching should emphasize reasoning from the category to multiple properties rather than from a set of properties to the category. (PsycINFO Database Record (c) 2010 APA, all rights reserved).

Machine learning algorithms for mode-of-action classification in toxicity assessment.

PubMed

Zhang, Yile; Wong, Yau Shu; Deng, Jian; Anton, Cristina; Gabos, Stephan; Zhang, Weiping; Huang, Dorothy Yu; Jin, Can

2016-01-01

Real Time Cell Analysis (RTCA) technology is used to monitor cellular changes continuously over the entire exposure period. Combining with different testing concentrations, the profiles have potential in probing the mode of action (MOA) of the testing substances. In this paper, we present machine learning approaches for MOA assessment. Computational tools based on artificial neural network (ANN) and support vector machine (SVM) are developed to analyze the time-concentration response curves (TCRCs) of human cell lines responding to tested chemicals. The techniques are capable of learning data from given TCRCs with known MOA information and then making MOA classification for the unknown toxicity. A novel data processing step based on wavelet transform is introduced to extract important features from the original TCRC data. From the dose response curves, time interval leading to higher classification success rate can be selected as input to enhance the performance of the machine learning algorithm. This is particularly helpful when handling cases with limited and imbalanced data. The validation of the proposed method is demonstrated by the supervised learning algorithm applied to the exposure data of HepG2 cell line to 63 chemicals with 11 concentrations in each test case. Classification success rate in the range of 85 to 95 % are obtained using SVM for MOA classification with two clusters to cases up to four clusters. Wavelet transform is capable of capturing important features of TCRCs for MOA classification. The proposed SVM scheme incorporated with wavelet transform has a great potential for large scale MOA classification and high-through output chemical screening.

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

PubMed Central

HUANG, SHUJUN; CAI, NIANGUANG; PACHECO, PEDRO PENZUTI; NARANDES, SHAVIRA; WANG, YANG; XU, WAYNE

2017-01-01

Machine learning with maximization (support) of separating margin (vector), called support vector machine (SVM) learning, is a powerful classification tool that has been used for cancer genomic classification or subtyping. Today, as advancements in high-throughput technologies lead to production of large amounts of genomic and epigenomic data, the classification feature of SVMs is expanding its use in cancer genomics, leading to the discovery of new biomarkers, new drug targets, and a better understanding of cancer driver genes. Herein we reviewed the recent progress of SVMs in cancer genomic studies. We intend to comprehend the strength of the SVM learning and its future perspective in cancer genomic applications. PMID:29275361

Intelligible machine learning with malibu.

PubMed

Langlois, Robert E; Lu, Hui

2008-01-01

malibu is an open-source machine learning work-bench developed in C/C++ for high-performance real-world applications, namely bioinformatics and medical informatics. It leverages third-party machine learning implementations for more robust bug-free software. This workbench handles several well-studied supervised machine learning problems including classification, regression, importance-weighted classification and multiple-instance learning. The malibu interface was designed to create reproducible experiments ideally run in a remote and/or command line environment. The software can be found at: http://proteomics.bioengr. uic.edu/malibu/index.html.

Sea ice classification using fast learning neural networks

NASA Technical Reports Server (NTRS)

Dawson, M. S.; Fung, A. K.; Manry, M. T.

1992-01-01

A first learning neural network approach to the classification of sea ice is presented. The fast learning (FL) neural network and a multilayer perceptron (MLP) trained with backpropagation learning (BP network) were tested on simulated data sets based on the known dominant scattering characteristics of the target class. Four classes were used in the data simulation: open water, thick lossy saline ice, thin saline ice, and multiyear ice. The BP network was unable to consistently converge to less than 25 percent error while the FL method yielded an average error of approximately 1 percent on the first iteration of training. The fast learning method presented can significantly reduce the CPU time necessary to train a neural network as well as consistently yield higher classification accuracy than BP networks.

Does Formative Assessment Improve Student Learning and Performance in Soil Science?

ERIC Educational Resources Information Center

Kopittke, Peter M.; Wehr, J. Bernhard; Menzies, Neal W.

2012-01-01

Soil science students are required to apply knowledge from a range of disciplines to unfamiliar scenarios to solve complex problems. To encourage deep learning (with student performance an indicator of learning), a formative assessment exercise was introduced to a second-year soil science subject. For the formative assessment exercise, students…

Semi-Supervised Marginal Fisher Analysis for Hyperspectral Image Classification

NASA Astrophysics Data System (ADS)

Huang, H.; Liu, J.; Pan, Y.

2012-07-01

The problem of learning with both labeled and unlabeled examples arises frequently in Hyperspectral image (HSI) classification. While marginal Fisher analysis is a supervised method, which cannot be directly applied for Semi-supervised classification. In this paper, we proposed a novel method, called semi-supervised marginal Fisher analysis (SSMFA), to process HSI of natural scenes, which uses a combination of semi-supervised learning and manifold learning. In SSMFA, a new difference-based optimization objective function with unlabeled samples has been designed. SSMFA preserves the manifold structure of labeled and unlabeled samples in addition to separating labeled samples in different classes from each other. The semi-supervised method has an analytic form of the globally optimal solution, and it can be computed based on eigen decomposition. Classification experiments with a challenging HSI task demonstrate that this method outperforms current state-of-the-art HSI-classification methods.

Manifold regularized multitask learning for semi-supervised multilabel image classification.

PubMed

Luo, Yong; Tao, Dacheng; Geng, Bo; Xu, Chao; Maybank, Stephen J

2013-02-01

It is a significant challenge to classify images with multiple labels by using only a small number of labeled samples. One option is to learn a binary classifier for each label and use manifold regularization to improve the classification performance by exploring the underlying geometric structure of the data distribution. However, such an approach does not perform well in practice when images from multiple concepts are represented by high-dimensional visual features. Thus, manifold regularization is insufficient to control the model complexity. In this paper, we propose a manifold regularized multitask learning (MRMTL) algorithm. MRMTL learns a discriminative subspace shared by multiple classification tasks by exploiting the common structure of these tasks. It effectively controls the model complexity because different tasks limit one another's search volume, and the manifold regularization ensures that the functions in the shared hypothesis space are smooth along the data manifold. We conduct extensive experiments, on the PASCAL VOC'07 dataset with 20 classes and the MIR dataset with 38 classes, by comparing MRMTL with popular image classification algorithms. The results suggest that MRMTL is effective for image classification.

Automated Classification of Radiology Reports for Acute Lung Injury: Comparison of Keyword and Machine Learning Based Natural Language Processing Approaches.

PubMed

Solti, Imre; Cooke, Colin R; Xia, Fei; Wurfel, Mark M

2009-11-01

This paper compares the performance of keyword and machine learning-based chest x-ray report classification for Acute Lung Injury (ALI). ALI mortality is approximately 30 percent. High mortality is, in part, a consequence of delayed manual chest x-ray classification. An automated system could reduce the time to recognize ALI and lead to reductions in mortality. For our study, 96 and 857 chest x-ray reports in two corpora were labeled by domain experts for ALI. We developed a keyword and a Maximum Entropy-based classification system. Word unigram and character n-grams provided the features for the machine learning system. The Maximum Entropy algorithm with character 6-gram achieved the highest performance (Recall=0.91, Precision=0.90 and F-measure=0.91) on the 857-report corpus. This study has shown that for the classification of ALI chest x-ray reports, the machine learning approach is superior to the keyword based system and achieves comparable results to highest performing physician annotators.

Automated Classification of Radiology Reports for Acute Lung Injury: Comparison of Keyword and Machine Learning Based Natural Language Processing Approaches

PubMed Central

Solti, Imre; Cooke, Colin R.; Xia, Fei; Wurfel, Mark M.

2010-01-01

This paper compares the performance of keyword and machine learning-based chest x-ray report classification for Acute Lung Injury (ALI). ALI mortality is approximately 30 percent. High mortality is, in part, a consequence of delayed manual chest x-ray classification. An automated system could reduce the time to recognize ALI and lead to reductions in mortality. For our study, 96 and 857 chest x-ray reports in two corpora were labeled by domain experts for ALI. We developed a keyword and a Maximum Entropy-based classification system. Word unigram and character n-grams provided the features for the machine learning system. The Maximum Entropy algorithm with character 6-gram achieved the highest performance (Recall=0.91, Precision=0.90 and F-measure=0.91) on the 857-report corpus. This study has shown that for the classification of ALI chest x-ray reports, the machine learning approach is superior to the keyword based system and achieves comparable results to highest performing physician annotators. PMID:21152268

Recent machine learning advancements in sensor-based mobility analysis: Deep learning for Parkinson's disease assessment.

PubMed

Eskofier, Bjoern M; Lee, Sunghoon I; Daneault, Jean-Francois; Golabchi, Fatemeh N; Ferreira-Carvalho, Gabriela; Vergara-Diaz, Gloria; Sapienza, Stefano; Costante, Gianluca; Klucken, Jochen; Kautz, Thomas; Bonato, Paolo

2016-08-01

The development of wearable sensors has opened the door for long-term assessment of movement disorders. However, there is still a need for developing methods suitable to monitor motor symptoms in and outside the clinic. The purpose of this paper was to investigate deep learning as a method for this monitoring. Deep learning recently broke records in speech and image classification, but it has not been fully investigated as a potential approach to analyze wearable sensor data. We collected data from ten patients with idiopathic Parkinson's disease using inertial measurement units. Several motor tasks were expert-labeled and used for classification. We specifically focused on the detection of bradykinesia. For this, we compared standard machine learning pipelines with deep learning based on convolutional neural networks. Our results showed that deep learning outperformed other state-of-the-art machine learning algorithms by at least 4.6 % in terms of classification rate. We contribute a discussion of the advantages and disadvantages of deep learning for sensor-based movement assessment and conclude that deep learning is a promising method for this field.

Soil-pipe interaction modeling for pipe behavior prediction with super learning based methods

NASA Astrophysics Data System (ADS)

Shi, Fang; Peng, Xiang; Liu, Huan; Hu, Yafei; Liu, Zheng; Li, Eric

2018-03-01

Underground pipelines are subject to severe distress from the surrounding expansive soil. To investigate the structural response of water mains to varying soil movements, field data, including pipe wall strains in situ soil water content, soil pressure and temperature, was collected. The research on monitoring data analysis has been reported, but the relationship between soil properties and pipe deformation has not been well-interpreted. To characterize the relationship between soil property and pipe deformation, this paper presents a super learning based approach combining feature selection algorithms to predict the water mains structural behavior in different soil environments. Furthermore, automatic variable selection method, e.i. recursive feature elimination algorithm, were used to identify the critical predictors contributing to the pipe deformations. To investigate the adaptability of super learning to different predictive models, this research employed super learning based methods to three different datasets. The predictive performance was evaluated by R-squared, root-mean-square error and mean absolute error. Based on the prediction performance evaluation, the superiority of super learning was validated and demonstrated by predicting three types of pipe deformations accurately. In addition, a comprehensive understand of the water mains working environments becomes possible.

Incidental Learning and Recall in Children.

ERIC Educational Resources Information Center

Fox, Robert A.; And Others

Incidental learning research with mentally retarded children has produced findings inconsistent with those reported for the intellectually normal population. This study was designed to further investigate the efficacy of incidental semantic classification instructions relative to taxonomic classification instructions or superficial color…

Reduction of Topographic Effect for Curve Number Estimated from Remotely Sensed Imagery

NASA Astrophysics Data System (ADS)

Zhang, Wen-Yan; Lin, Chao-Yuan

2016-04-01

The Soil Conservation Service Curve Number (SCS-CN) method is commonly used in hydrology to estimate direct runoff volume. The CN is the empirical parameter which corresponding to land use/land cover, hydrologic soil group and antecedent soil moisture condition. In large watersheds with complex topography, satellite remote sensing is the appropriate approach to acquire the land use change information. However, the topographic effect have been usually found in the remotely sensed imageries and resulted in land use classification. This research selected summer and winter scenes of Landsat-5 TM during 2008 to classified land use in Chen-You-Lan Watershed, Taiwan. The b-correction, the empirical topographic correction method, was applied to Landsat-5 TM data. Land use were categorized using K-mean classification into 4 groups i.e. forest, grassland, agriculture and river. Accuracy assessment of image classification was performed with national land use map. The results showed that after topographic correction, the overall accuracy of classification was increased from 68.0% to 74.5%. The average CN estimated from remotely sensed imagery decreased from 48.69 to 45.35 where the average CN estimated from national LULC map was 44.11. Therefore, the topographic correction method was recommended to normalize the topographic effect from the satellite remote sensing data before estimating the CN.

Recognizing Banknote Fitness with a Visible Light One Dimensional Line Image Sensor

PubMed Central

Pham, Tuyen Danh; Park, Young Ho; Kwon, Seung Yong; Nguyen, Dat Tien; Vokhidov, Husan; Park, Kang Ryoung; Jeong, Dae Sik; Yoon, Sungsoo

2015-01-01

In general, dirty banknotes that have creases or soiled surfaces should be replaced by new banknotes, whereas clean banknotes should be recirculated. Therefore, the accurate classification of banknote fitness when sorting paper currency is an important and challenging task. Most previous research has focused on sensors that used visible, infrared, and ultraviolet light. Furthermore, there was little previous research on the fitness classification for Indian paper currency. Therefore, we propose a new method for classifying the fitness of Indian banknotes, with a one-dimensional line image sensor that uses only visible light. The fitness of banknotes is usually determined by various factors such as soiling, creases, and tears, etc. although we just consider banknote soiling in our research. This research is novel in the following four ways: first, there has been little research conducted on fitness classification for the Indian Rupee using visible-light images. Second, the classification is conducted based on the features extracted from the regions of interest (ROIs), which contain little texture. Third, 1-level discrete wavelet transformation (DWT) is used to extract the features for discriminating between fit and unfit banknotes. Fourth, the optimal DWT features that represent the fitness and unfitness of banknotes are selected based on linear regression analysis with ground-truth data measured by densitometer. In addition, the selected features are used as the inputs to a support vector machine (SVM) for the final classification of banknote fitness. Experimental results showed that our method outperforms other methods. PMID:26343654

Predictive mapping of soil organic carbon in wet cultivated lands using classification-tree based models: the case study of Denmark.

PubMed

Bou Kheir, Rania; Greve, Mogens H; Bøcher, Peder K; Greve, Mette B; Larsen, René; McCloy, Keith

2010-05-01

Soil organic carbon (SOC) is one of the most important carbon stocks globally and has large potential to affect global climate. Distribution patterns of SOC in Denmark constitute a nation-wide baseline for studies on soil carbon changes (with respect to Kyoto protocol). This paper predicts and maps the geographic distribution of SOC across Denmark using remote sensing (RS), geographic information systems (GISs) and decision-tree modeling (un-pruned and pruned classification trees). Seventeen parameters, i.e. parent material, soil type, landscape type, elevation, slope gradient, slope aspect, mean curvature, plan curvature, profile curvature, flow accumulation, specific catchment area, tangent slope, tangent curvature, steady-state wetness index, Normalized Difference Vegetation Index (NDVI), Normalized Difference Wetness Index (NDWI) and Soil Color Index (SCI) were generated to statistically explain SOC field measurements in the area of interest (Denmark). A large number of tree-based classification models (588) were developed using (i) all of the parameters, (ii) all Digital Elevation Model (DEM) parameters only, (iii) the primary DEM parameters only, (iv), the remote sensing (RS) indices only, (v) selected pairs of parameters, (vi) soil type, parent material and landscape type only, and (vii) the parameters having a high impact on SOC distribution in built pruned trees. The best constructed classification tree models (in the number of three) with the lowest misclassification error (ME) and the lowest number of nodes (N) as well are: (i) the tree (T1) combining all of the parameters (ME=29.5%; N=54); (ii) the tree (T2) based on the parent material, soil type and landscape type (ME=31.5%; N=14); and (iii) the tree (T3) constructed using parent material, soil type, landscape type, elevation, tangent slope and SCI (ME=30%; N=39). The produced SOC maps at 1:50,000 cartographic scale using these trees are highly matching with coincidence values equal to 90.5% (Map T1/Map T2), 95% (Map T1/Map T3) and 91% (Map T2/Map T3). The overall accuracies of these maps once compared with field observations were estimated to be 69.54% (Map T1), 68.87% (Map T2) and 69.41% (Map T3). The proposed tree models are relatively simple, and may be also applied to other areas. Copyright 2010 Elsevier Ltd. All rights reserved.

Optimized extreme learning machine for urban land cover classification using hyperspectral imagery

NASA Astrophysics Data System (ADS)

Su, Hongjun; Tian, Shufang; Cai, Yue; Sheng, Yehua; Chen, Chen; Najafian, Maryam

2017-12-01

This work presents a new urban land cover classification framework using the firefly algorithm (FA) optimized extreme learning machine (ELM). FA is adopted to optimize the regularization coefficient C and Gaussian kernel σ for kernel ELM. Additionally, effectiveness of spectral features derived from an FA-based band selection algorithm is studied for the proposed classification task. Three sets of hyperspectral databases were recorded using different sensors, namely HYDICE, HyMap, and AVIRIS. Our study shows that the proposed method outperforms traditional classification algorithms such as SVM and reduces computational cost significantly.

Classification and evaluation for forest sites on the Western Highland Rim and Pennyroyal

Treesearch

Glendon W. Smalley

1980-01-01

Presents a comprehensive forest site classification system for the Western Highland Rim and Western Pennyroyal-Limestone area in northwest Alabama, west-central Tennessee, and western Kentucky. The system is based on physiography, geology, soils, topography, and vegetation.

Auto-SEIA: simultaneous optimization of image processing and machine learning algorithms

NASA Astrophysics Data System (ADS)

Negro Maggio, Valentina; Iocchi, Luca

2015-02-01

Object classification from images is an important task for machine vision and it is a crucial ingredient for many computer vision applications, ranging from security and surveillance to marketing. Image based object classification techniques properly integrate image processing and machine learning (i.e., classification) procedures. In this paper we present a system for automatic simultaneous optimization of algorithms and parameters for object classification from images. More specifically, the proposed system is able to process a dataset of labelled images and to return a best configuration of image processing and classification algorithms and of their parameters with respect to the accuracy of classification. Experiments with real public datasets are used to demonstrate the effectiveness of the developed system.

An integrated Landsat/ancillary data classification of desert rangeland

NASA Technical Reports Server (NTRS)

Price, K. P.; Ridd, M. K.; Merola, J. A.

1985-01-01

Range inventorying methods using Landsat MSS data, coupled with ancillary data were examined. The study area encompassed nearly 20,000 acres in Rush Valley, UT. The vegetation is predominately desert shrub and annual grasses, with same annual forbs. Three Landsat scenes were evaluated using a Kauth-Thomas brightness/greenness data transformation (May, June, and August dates). The data was classified using a four-band maximum-likelihood classifier. A print map was taken into the field to determine the relationship between print symbols and vegetation. It was determined that classification confusion could be greatly reduced by incorporating geomorphic units and soil texture (coarse vs fine) into the classification. Spectral data, geomorphic units, and soil texture were combined in a GIS format to produce a final vegetation map identifying 12 vegetation types.

An integrated LANDSAT/ancillary data classification of desert rangeland

NASA Technical Reports Server (NTRS)

Price, K. P.; Ridd, M. K.; Merola, J. A.

1984-01-01

Range inventorying methods using LANDSAT MSS data, coupled with ancillary data were examined. The study area encompassed nearly 20,000 acres in Rush Valley, Utah. The vegetation is predominately desert shrub and annual grasses, with some annual forbs. Three LANDSAT scenes were evaluated using a Kauth-Thomas brightness/greenness data transformation (May, June, and August dates). The data was classified using a four-band maximum-likelihood classifier. A print map was taken into the field to determine the relationship between print symbols and vegetation. It was determined that classification confusion could be greatly reduced by incorporating geomorphic units and soil texture (coarse vs fine) into the classification. Spectral data, geomorphic units, and soil texture were combined in a GIS format to produce a final vegetation map identifying 12 vegetation types.

Color Image Classification Using Block Matching and Learning

NASA Astrophysics Data System (ADS)

Kondo, Kazuki; Hotta, Seiji

In this paper, we propose block matching and learning for color image classification. In our method, training images are partitioned into small blocks. Given a test image, it is also partitioned into small blocks, and mean-blocks corresponding to each test block are calculated with neighbor training blocks. Our method classifies a test image into the class that has the shortest total sum of distances between mean blocks and test ones. We also propose a learning method for reducing memory requirement. Experimental results show that our classification outperforms other classifiers such as support vector machine with bag of keypoints.

Use of machine learning methods to classify Universities based on the income structure

NASA Astrophysics Data System (ADS)

Terlyga, Alexandra; Balk, Igor

2017-10-01

In this paper we discuss use of machine learning methods such as self organizing maps, k-means and Ward’s clustering to perform classification of universities based on their income. This classification will allow us to quantitate classification of universities as teaching, research, entrepreneur, etc. which is important tool for government, corporations and general public alike in setting expectation and selecting universities to achieve different goals.

Use of an automatic procedure for determination of classes of land use in the Teste Araras area of the peripheral Paulist depression

NASA Technical Reports Server (NTRS)

Dejesusparada, N. (Principal Investigator); Lombardo, M. A.; Valeriano, D. D.

1981-01-01

An evaluation of the multispectral image analyzer (system Image 1-100), using automatic classification, is presented. The region studied is situated. The automatic was carried out using the maximum likelihood (MAXVER) classification system. The following classes were established: urban area, bare soil, sugar cane, citrus culture (oranges), pastures, and reforestation. The classification matrix of the test sites indicate that the percentage of correct classification varied between 63% and 100%.

Multimodal Task-Driven Dictionary Learning for Image Classification

DTIC Science & Technology

2015-12-18

1 Multimodal Task-Driven Dictionary Learning for Image Classification Soheil Bahrampour, Student Member, IEEE, Nasser M. Nasrabadi, Fellow, IEEE...Asok Ray, Fellow, IEEE, and W. Kenneth Jenkins, Life Fellow, IEEE Abstract— Dictionary learning algorithms have been suc- cessfully used for both...reconstructive and discriminative tasks, where an input signal is represented with a sparse linear combination of dictionary atoms. While these methods are

Machine learning approach for automated screening of malaria parasite using light microscopic images.

PubMed

Das, Dev Kumar; Ghosh, Madhumala; Pal, Mallika; Maiti, Asok K; Chakraborty, Chandan

2013-02-01

The aim of this paper is to address the development of computer assisted malaria parasite characterization and classification using machine learning approach based on light microscopic images of peripheral blood smears. In doing this, microscopic image acquisition from stained slides, illumination correction and noise reduction, erythrocyte segmentation, feature extraction, feature selection and finally classification of different stages of malaria (Plasmodium vivax and Plasmodium falciparum) have been investigated. The erythrocytes are segmented using marker controlled watershed transformation and subsequently total ninety six features describing shape-size and texture of erythrocytes are extracted in respect to the parasitemia infected versus non-infected cells. Ninety four features are found to be statistically significant in discriminating six classes. Here a feature selection-cum-classification scheme has been devised by combining F-statistic, statistical learning techniques i.e., Bayesian learning and support vector machine (SVM) in order to provide the higher classification accuracy using best set of discriminating features. Results show that Bayesian approach provides the highest accuracy i.e., 84% for malaria classification by selecting 19 most significant features while SVM provides highest accuracy i.e., 83.5% with 9 most significant features. Finally, the performance of these two classifiers under feature selection framework has been compared toward malaria parasite classification. Copyright © 2012 Elsevier Ltd. All rights reserved.

Advanced Steel Microstructural Classification by Deep Learning Methods.

PubMed

Azimi, Seyed Majid; Britz, Dominik; Engstler, Michael; Fritz, Mario; Mücklich, Frank

2018-02-01

The inner structure of a material is called microstructure. It stores the genesis of a material and determines all its physical and chemical properties. While microstructural characterization is widely spread and well known, the microstructural classification is mostly done manually by human experts, which gives rise to uncertainties due to subjectivity. Since the microstructure could be a combination of different phases or constituents with complex substructures its automatic classification is very challenging and only a few prior studies exist. Prior works focused on designed and engineered features by experts and classified microstructures separately from the feature extraction step. Recently, Deep Learning methods have shown strong performance in vision applications by learning the features from data together with the classification step. In this work, we propose a Deep Learning method for microstructural classification in the examples of certain microstructural constituents of low carbon steel. This novel method employs pixel-wise segmentation via Fully Convolutional Neural Network (FCNN) accompanied by a max-voting scheme. Our system achieves 93.94% classification accuracy, drastically outperforming the state-of-the-art method of 48.89% accuracy. Beyond the strong performance of our method, this line of research offers a more robust and first of all objective way for the difficult task of steel quality appreciation.

Integrated feature extraction and selection for neuroimage classification

NASA Astrophysics Data System (ADS)

Fan, Yong; Shen, Dinggang

2009-02-01

Feature extraction and selection are of great importance in neuroimage classification for identifying informative features and reducing feature dimensionality, which are generally implemented as two separate steps. This paper presents an integrated feature extraction and selection algorithm with two iterative steps: constrained subspace learning based feature extraction and support vector machine (SVM) based feature selection. The subspace learning based feature extraction focuses on the brain regions with higher possibility of being affected by the disease under study, while the possibility of brain regions being affected by disease is estimated by the SVM based feature selection, in conjunction with SVM classification. This algorithm can not only take into account the inter-correlation among different brain regions, but also overcome the limitation of traditional subspace learning based feature extraction methods. To achieve robust performance and optimal selection of parameters involved in feature extraction, selection, and classification, a bootstrapping strategy is used to generate multiple versions of training and testing sets for parameter optimization, according to the classification performance measured by the area under the ROC (receiver operating characteristic) curve. The integrated feature extraction and selection method is applied to a structural MR image based Alzheimer's disease (AD) study with 98 non-demented and 100 demented subjects. Cross-validation results indicate that the proposed algorithm can improve performance of the traditional subspace learning based classification.

Classification of Strawberry Fruit Shape by Machine Learning

NASA Astrophysics Data System (ADS)

Ishikawa, T.; Hayashi, A.; Nagamatsu, S.; Kyutoku, Y.; Dan, I.; Wada, T.; Oku, K.; Saeki, Y.; Uto, T.; Tanabata, T.; Isobe, S.; Kochi, N.

2018-05-01

Shape is one of the most important traits of agricultural products due to its relationships with the quality, quantity, and value of the products. For strawberries, the nine types of fruit shape were defined and classified by humans based on the sampler patterns of the nine types. In this study, we tested the classification of strawberry shapes by machine learning in order to increase the accuracy of the classification, and we introduce the concept of computerization into this field. Four types of descriptors were extracted from the digital images of strawberries: (1) the Measured Values (MVs) including the length of the contour line, the area, the fruit length and width, and the fruit width/length ratio; (2) the Ellipse Similarity Index (ESI); (3) Elliptic Fourier Descriptors (EFDs), and (4) Chain Code Subtraction (CCS). We used these descriptors for the classification test along with the random forest approach, and eight of the nine shape types were classified with combinations of MVs + CCS + EFDs. CCS is a descriptor that adds human knowledge to the chain codes, and it showed higher robustness in classification than the other descriptors. Our results suggest machine learning's high ability to classify fruit shapes accurately. We will attempt to increase the classification accuracy and apply the machine learning methods to other plant species.

Unsupervised active learning based on hierarchical graph-theoretic clustering.

PubMed

Hu, Weiming; Hu, Wei; Xie, Nianhua; Maybank, Steve

2009-10-01

Most existing active learning approaches are supervised. Supervised active learning has the following problems: inefficiency in dealing with the semantic gap between the distribution of samples in the feature space and their labels, lack of ability in selecting new samples that belong to new categories that have not yet appeared in the training samples, and lack of adaptability to changes in the semantic interpretation of sample categories. To tackle these problems, we propose an unsupervised active learning framework based on hierarchical graph-theoretic clustering. In the framework, two promising graph-theoretic clustering algorithms, namely, dominant-set clustering and spectral clustering, are combined in a hierarchical fashion. Our framework has some advantages, such as ease of implementation, flexibility in architecture, and adaptability to changes in the labeling. Evaluations on data sets for network intrusion detection, image classification, and video classification have demonstrated that our active learning framework can effectively reduce the workload of manual classification while maintaining a high accuracy of automatic classification. It is shown that, overall, our framework outperforms the support-vector-machine-based supervised active learning, particularly in terms of dealing much more efficiently with new samples whose categories have not yet appeared in the training samples.

Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments.

PubMed

Han, Wenjing; Coutinho, Eduardo; Ruan, Huabin; Li, Haifeng; Schuller, Björn; Yu, Xiaojie; Zhu, Xuan

2016-01-01

Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an efficient combination of confidence-based Active Learning and Self-Training with the aim of minimizing the need for human annotation for sound classification model training. The proposed method pre-processes the instances that are ready for labeling by calculating their classifier confidence scores, and then delivers the candidates with lower scores to human annotators, and those with high scores are automatically labeled by the machine. We demonstrate the feasibility and efficacy of this method in two practical scenarios: pool-based and stream-based processing. Extensive experimental results indicate that our approach requires significantly less labeled instances to reach the same performance in both scenarios compared to Passive Learning, Active Learning and Self-Training. A reduction of 52.2% in human labeled instances is achieved in both of the pool-based and stream-based scenarios on a sound classification task considering 16,930 sound instances.

Semi-supervised manifold learning with affinity regularization for Alzheimer's disease identification using positron emission tomography imaging.

PubMed

Lu, Shen; Xia, Yong; Cai, Tom Weidong; Feng, David Dagan

2015-01-01

Dementia, Alzheimer's disease (AD) in particular is a global problem and big threat to the aging population. An image based computer-aided dementia diagnosis method is needed to providing doctors help during medical image examination. Many machine learning based dementia classification methods using medical imaging have been proposed and most of them achieve accurate results. However, most of these methods make use of supervised learning requiring fully labeled image dataset, which usually is not practical in real clinical environment. Using large amount of unlabeled images can improve the dementia classification performance. In this study we propose a new semi-supervised dementia classification method based on random manifold learning with affinity regularization. Three groups of spatial features are extracted from positron emission tomography (PET) images to construct an unsupervised random forest which is then used to regularize the manifold learning objective function. The proposed method, stat-of-the-art Laplacian support vector machine (LapSVM) and supervised SVM are applied to classify AD and normal controls (NC). The experiment results show that learning with unlabeled images indeed improves the classification performance. And our method outperforms LapSVM on the same dataset.

A Novel Extreme Learning Machine Classification Model for e-Nose Application Based on the Multiple Kernel Approach.

PubMed

Jian, Yulin; Huang, Daoyu; Yan, Jia; Lu, Kun; Huang, Ying; Wen, Tailai; Zeng, Tanyue; Zhong, Shijie; Xie, Qilong

2017-06-19

A novel classification model, named the quantum-behaved particle swarm optimization (QPSO)-based weighted multiple kernel extreme learning machine (QWMK-ELM), is proposed in this paper. Experimental validation is carried out with two different electronic nose (e-nose) datasets. Being different from the existing multiple kernel extreme learning machine (MK-ELM) algorithms, the combination coefficients of base kernels are regarded as external parameters of single-hidden layer feedforward neural networks (SLFNs). The combination coefficients of base kernels, the model parameters of each base kernel, and the regularization parameter are optimized by QPSO simultaneously before implementing the kernel extreme learning machine (KELM) with the composite kernel function. Four types of common single kernel functions (Gaussian kernel, polynomial kernel, sigmoid kernel, and wavelet kernel) are utilized to constitute different composite kernel functions. Moreover, the method is also compared with other existing classification methods: extreme learning machine (ELM), kernel extreme learning machine (KELM), k-nearest neighbors (KNN), support vector machine (SVM), multi-layer perceptron (MLP), radical basis function neural network (RBFNN), and probabilistic neural network (PNN). The results have demonstrated that the proposed QWMK-ELM outperforms the aforementioned methods, not only in precision, but also in efficiency for gas classification.

Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments

PubMed Central

Han, Wenjing; Coutinho, Eduardo; Li, Haifeng; Schuller, Björn; Yu, Xiaojie; Zhu, Xuan

2016-01-01

Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an efficient combination of confidence-based Active Learning and Self-Training with the aim of minimizing the need for human annotation for sound classification model training. The proposed method pre-processes the instances that are ready for labeling by calculating their classifier confidence scores, and then delivers the candidates with lower scores to human annotators, and those with high scores are automatically labeled by the machine. We demonstrate the feasibility and efficacy of this method in two practical scenarios: pool-based and stream-based processing. Extensive experimental results indicate that our approach requires significantly less labeled instances to reach the same performance in both scenarios compared to Passive Learning, Active Learning and Self-Training. A reduction of 52.2% in human labeled instances is achieved in both of the pool-based and stream-based scenarios on a sound classification task considering 16,930 sound instances. PMID:27627768

The stability of clay using mount Sinabung ash with unconfined compression test (uct) value

NASA Astrophysics Data System (ADS)

Puji Hastuty, Ika; Roesyanto; Hutauruk, Ronny; Simanjuntak, Oberlyn

2018-03-01

The soil has a important role as a highway’s embankment material (sub grade). Soil conditions are very different in each location because the scientifically soil is a very complex and varied material and the located on the field is very loose or very soft, so it is not suitable for construction, then the soil should be stabilized. The additive material commonly used for soil stabilization includes cement, lime, fly ash, rice husk ash, and others. This experiment is using the addition of volcanic ash. The purpose of this study was to determine the Index Properties and Compressive Strength maximum value with Unconfined Compression Test due to the addition of volcanic ash as a stabilizing agent along with optimum levels of the addition. The result showed that the original soil sample has Water Content of 14.52%; the Specific Weight of 2.64%; Liquid limit of 48.64% and Plasticity Index of 29.82%. Then, the Compressive Strength value is 1.40 kg/cm2. According to USCS classification, the soil samples categorized as the (CL) type while based on AASHTO classification, the soil samples are including as the type of A-7-6. After the soil is stabilized with a variety of volcanic ash, can be concluded that the maximum value occurs at mixture variation of 11% Volcanic Ash with Unconfined Compressive Strength value of 2.32 kg/cm2.

Seizure Classification From EEG Signals Using Transfer Learning, Semi-Supervised Learning and TSK Fuzzy System.

PubMed

Jiang, Yizhang; Wu, Dongrui; Deng, Zhaohong; Qian, Pengjiang; Wang, Jun; Wang, Guanjin; Chung, Fu-Lai; Choi, Kup-Sze; Wang, Shitong

2017-12-01

Recognition of epileptic seizures from offline EEG signals is very important in clinical diagnosis of epilepsy. Compared with manual labeling of EEG signals by doctors, machine learning approaches can be faster and more consistent. However, the classification accuracy is usually not satisfactory for two main reasons: the distributions of the data used for training and testing may be different, and the amount of training data may not be enough. In addition, most machine learning approaches generate black-box models that are difficult to interpret. In this paper, we integrate transductive transfer learning, semi-supervised learning and TSK fuzzy system to tackle these three problems. More specifically, we use transfer learning to reduce the discrepancy in data distribution between the training and testing data, employ semi-supervised learning to use the unlabeled testing data to remedy the shortage of training data, and adopt TSK fuzzy system to increase model interpretability. Two learning algorithms are proposed to train the system. Our experimental results show that the proposed approaches can achieve better performance than many state-of-the-art seizure classification algorithms.

A novel application of deep learning for single-lead ECG classification.

PubMed

Mathews, Sherin M; Kambhamettu, Chandra; Barner, Kenneth E

2018-06-04

Detecting and classifying cardiac arrhythmias is critical to the diagnosis of patients with cardiac abnormalities. In this paper, a novel approach based on deep learning methodology is proposed for the classification of single-lead electrocardiogram (ECG) signals. We demonstrate the application of the Restricted Boltzmann Machine (RBM) and deep belief networks (DBN) for ECG classification following detection of ventricular and supraventricular heartbeats using single-lead ECG. The effectiveness of this proposed algorithm is illustrated using real ECG signals from the widely-used MIT-BIH database. Simulation results demonstrate that with a suitable choice of parameters, RBM and DBN can achieve high average recognition accuracies of ventricular ectopic beats (93.63%) and of supraventricular ectopic beats (95.57%) at a low sampling rate of 114 Hz. Experimental results indicate that classifiers built into this deep learning-based framework achieved state-of-the art performance models at lower sampling rates and simple features when compared to traditional methods. Further, employing features extracted at a sampling rate of 114 Hz when combined with deep learning provided enough discriminatory power for the classification task. This performance is comparable to that of traditional methods and uses a much lower sampling rate and simpler features. Thus, our proposed deep neural network algorithm demonstrates that deep learning-based methods offer accurate ECG classification and could potentially be extended to other physiological signal classifications, such as those in arterial blood pressure (ABP), nerve conduction (EMG), and heart rate variability (HRV) studies. Copyright © 2018. Published by Elsevier Ltd.

Deep learning for EEG-Based preference classification

NASA Astrophysics Data System (ADS)

Teo, Jason; Hou, Chew Lin; Mountstephens, James

2017-10-01

Electroencephalogram (EEG)-based emotion classification is rapidly becoming one of the most intensely studied areas of brain-computer interfacing (BCI). The ability to passively identify yet accurately correlate brainwaves with our immediate emotions opens up truly meaningful and previously unattainable human-computer interactions such as in forensic neuroscience, rehabilitative medicine, affective entertainment and neuro-marketing. One particularly useful yet rarely explored areas of EEG-based emotion classification is preference recognition [1], which is simply the detection of like versus dislike. Within the limited investigations into preference classification, all reported studies were based on musically-induced stimuli except for a single study which used 2D images. The main objective of this study is to apply deep learning, which has been shown to produce state-of-the-art results in diverse hard problems such as in computer vision, natural language processing and audio recognition, to 3D object preference classification over a larger group of test subjects. A cohort of 16 users was shown 60 bracelet-like objects as rotating visual stimuli on a computer display while their preferences and EEGs were recorded. After training a variety of machine learning approaches which included deep neural networks, we then attempted to classify the users' preferences for the 3D visual stimuli based on their EEGs. Here, we show that that deep learning outperforms a variety of other machine learning classifiers for this EEG-based preference classification task particularly in a highly challenging dataset with large inter- and intra-subject variability.

Manifold Regularized Experimental Design for Active Learning.

PubMed

Zhang, Lining; Shum, Hubert P H; Shao, Ling

2016-12-02

Various machine learning and data mining tasks in classification require abundant data samples to be labeled for training. Conventional active learning methods aim at labeling the most informative samples for alleviating the labor of the user. Many previous studies in active learning select one sample after another in a greedy manner. However, this is not very effective because the classification models has to be retrained for each newly labeled sample. Moreover, many popular active learning approaches utilize the most uncertain samples by leveraging the classification hyperplane of the classifier, which is not appropriate since the classification hyperplane is inaccurate when the training data are small-sized. The problem of insufficient training data in real-world systems limits the potential applications of these approaches. This paper presents a novel method of active learning called manifold regularized experimental design (MRED), which can label multiple informative samples at one time for training. In addition, MRED gives an explicit geometric explanation for the selected samples to be labeled by the user. Different from existing active learning methods, our method avoids the intrinsic problems caused by insufficiently labeled samples in real-world applications. Various experiments on synthetic datasets, the Yale face database and the Corel image database have been carried out to show how MRED outperforms existing methods.

Automatic Estimation of Osteoporotic Fracture Cases by Using Ensemble Learning Approaches.

PubMed

Kilic, Niyazi; Hosgormez, Erkan

2016-03-01

Ensemble learning methods are one of the most powerful tools for the pattern classification problems. In this paper, the effects of ensemble learning methods and some physical bone densitometry parameters on osteoporotic fracture detection were investigated. Six feature set models were constructed including different physical parameters and they fed into the ensemble classifiers as input features. As ensemble learning techniques, bagging, gradient boosting and random subspace (RSM) were used. Instance based learning (IBk) and random forest (RF) classifiers applied to six feature set models. The patients were classified into three groups such as osteoporosis, osteopenia and control (healthy), using ensemble classifiers. Total classification accuracy and f-measure were also used to evaluate diagnostic performance of the proposed ensemble classification system. The classification accuracy has reached to 98.85 % by the combination of model 6 (five BMD + five T-score values) using RSM-RF classifier. The findings of this paper suggest that the patients will be able to be warned before a bone fracture occurred, by just examining some physical parameters that can easily be measured without invasive operations.

Attention Recognition in EEG-Based Affective Learning Research Using CFS+KNN Algorithm.

PubMed

Hu, Bin; Li, Xiaowei; Sun, Shuting; Ratcliffe, Martyn

2018-01-01

The research detailed in this paper focuses on the processing of Electroencephalography (EEG) data to identify attention during the learning process. The identification of affect using our procedures is integrated into a simulated distance learning system that provides feedback to the user with respect to attention and concentration. The authors propose a classification procedure that combines correlation-based feature selection (CFS) and a k-nearest-neighbor (KNN) data mining algorithm. To evaluate the CFS+KNN algorithm, it was test against CFS+C4.5 algorithm and other classification algorithms. The classification performance was measured 10 times with different 3-fold cross validation data. The data was derived from 10 subjects while they were attempting to learn material in a simulated distance learning environment. A self-assessment model of self-report was used with a single valence to evaluate attention on 3 levels (high, neutral, low). It was found that CFS+KNN had a much better performance, giving the highest correct classification rate (CCR) of % for the valence dimension divided into three classes.

Optimizing support vector machine learning for semi-arid vegetation mapping by using clustering analysis

NASA Astrophysics Data System (ADS)

Su, Lihong

In remote sensing communities, support vector machine (SVM) learning has recently received increasing attention. SVM learning usually requires large memory and enormous amounts of computation time on large training sets. According to SVM algorithms, the SVM classification decision function is fully determined by support vectors, which compose a subset of the training sets. In this regard, a solution to optimize SVM learning is to efficiently reduce training sets. In this paper, a data reduction method based on agglomerative hierarchical clustering is proposed to obtain smaller training sets for SVM learning. Using a multiple angle remote sensing dataset of a semi-arid region, the effectiveness of the proposed method is evaluated by classification experiments with a series of reduced training sets. The experiments show that there is no loss of SVM accuracy when the original training set is reduced to 34% using the proposed approach. Maximum likelihood classification (MLC) also is applied on the reduced training sets. The results show that MLC can also maintain the classification accuracy. This implies that the most informative data instances can be retained by this approach.

Semi-Supervised Projective Non-Negative Matrix Factorization for Cancer Classification.

PubMed

Zhang, Xiang; Guan, Naiyang; Jia, Zhilong; Qiu, Xiaogang; Luo, Zhigang

2015-01-01

Advances in DNA microarray technologies have made gene expression profiles a significant candidate in identifying different types of cancers. Traditional learning-based cancer identification methods utilize labeled samples to train a classifier, but they are inconvenient for practical application because labels are quite expensive in the clinical cancer research community. This paper proposes a semi-supervised projective non-negative matrix factorization method (Semi-PNMF) to learn an effective classifier from both labeled and unlabeled samples, thus boosting subsequent cancer classification performance. In particular, Semi-PNMF jointly learns a non-negative subspace from concatenated labeled and unlabeled samples and indicates classes by the positions of the maximum entries of their coefficients. Because Semi-PNMF incorporates statistical information from the large volume of unlabeled samples in the learned subspace, it can learn more representative subspaces and boost classification performance. We developed a multiplicative update rule (MUR) to optimize Semi-PNMF and proved its convergence. The experimental results of cancer classification for two multiclass cancer gene expression profile datasets show that Semi-PNMF outperforms the representative methods.

Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks.

PubMed

Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R; Nguyen, Tuan N; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T

2017-01-01

This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively.

Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks

PubMed Central

Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R.; Nguyen, Tuan N.; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T.

2017-01-01

This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively. PMID:28326009

Network-based high level data classification.

PubMed

Silva, Thiago Christiano; Zhao, Liang

2012-06-01

Traditional supervised data classification considers only physical features (e.g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Differential impact of relevant and irrelevant dimension primes on rule-based and information-integration category learning.

PubMed

Grimm, Lisa R; Maddox, W Todd

2013-11-01

Research has identified multiple category-learning systems with each being "tuned" for learning categories with different task demands and each governed by different neurobiological systems. Rule-based (RB) classification involves testing verbalizable rules for category membership while information-integration (II) classification requires the implicit learning of stimulus-response mappings. In the first study to directly test rule priming with RB and II category learning, we investigated the influence of the availability of information presented at the beginning of the task. Participants viewed lines that varied in length, orientation, and position on the screen, and were primed to focus on stimulus dimensions that were relevant or irrelevant to the correct classification rule. In Experiment 1, we used an RB category structure, and in Experiment 2, we used an II category structure. Accuracy and model-based analyses suggested that a focus on relevant dimensions improves RB task performance later in learning while a focus on an irrelevant dimension improves II task performance early in learning. © 2013.

Helping People Understand Soils - Perspectives from the US National Cooperative Soil Survey

NASA Astrophysics Data System (ADS)

Reich, Paul; Cheever, Tammy; Greene, Linda; Southard, Susan; Levin, Maxine; Lindbo, David L.; Monger, Curtis

2017-04-01

Throughout the history of the US National Cooperative Soil Survey (NCSS), soil science education has been a part of the mission to better understand one of our most precious natural resources: the Soil. The poster will highlight the many products and programs related to soils that USDA NRCS (soils.usda.gov) has developed over the years for K-12 and college/professional education. NRCS scientific publications covering topics on soil properties, soil classification, soil health and soil quality have become an important part of the university soil science curriculum. Classroom lesson plans and grade appropriate materials help K-12 teachers introduce soil concepts to students and include detailed instructions and materials for classroom demonstrations of soil properties. A Handbook for Collegiate Soils Contests support universities that conduct Collegiate Soil Judging contests.

A comparative study for chest radiograph image retrieval using binary texture and deep learning classification.

PubMed

Anavi, Yaron; Kogan, Ilya; Gelbart, Elad; Geva, Ofer; Greenspan, Hayit

2015-08-01

In this work various approaches are investigated for X-ray image retrieval and specifically chest pathology retrieval. Given a query image taken from a data set of 443 images, the objective is to rank images according to similarity. Different features, including binary features, texture features, and deep learning (CNN) features are examined. In addition, two approaches are investigated for the retrieval task. One approach is based on the distance of image descriptors using the above features (hereon termed the "descriptor"-based approach); the second approach ("classification"-based approach) is based on a probability descriptor, generated by a pair-wise classification of each two classes (pathologies) and their decision values using an SVM classifier. Best results are achieved using deep learning features in a classification scheme.

exprso: an R-package for the rapid implementation of machine learning algorithms.

PubMed

Quinn, Thomas; Tylee, Daniel; Glatt, Stephen

2016-01-01

Machine learning plays a major role in many scientific investigations. However, non-expert programmers may struggle to implement the elaborate pipelines necessary to build highly accurate and generalizable models. We introduce exprso , a new R package that is an intuitive machine learning suite designed specifically for non-expert programmers. Built initially for the classification of high-dimensional data, exprso uses an object-oriented framework to encapsulate a number of common analytical methods into a series of interchangeable modules. This includes modules for feature selection, classification, high-throughput parameter grid-searching, elaborate cross-validation schemes (e.g., Monte Carlo and nested cross-validation), ensemble classification, and prediction. In addition, exprso also supports multi-class classification (through the 1-vs-all generalization of binary classifiers) and the prediction of continuous outcomes.

Classification

NASA Astrophysics Data System (ADS)

Oza, Nikunj

2012-03-01

A supervised learning task involves constructing a mapping from input data (normally described by several features) to the appropriate outputs. A set of training examples— examples with known output values—is used by a learning algorithm to generate a model. This model is intended to approximate the mapping between the inputs and outputs. This model can be used to generate predicted outputs for inputs that have not been seen before. Within supervised learning, one type of task is a classification learning task, in which each output is one or more classes to which the input belongs. For example, we may have data consisting of observations of sunspots. In a classification learning task, our goal may be to learn to classify sunspots into one of several types. Each example may correspond to one candidate sunspot with various measurements or just an image. A learning algorithm would use the supplied examples to generate a model that approximates the mapping between each supplied set of measurements and the type of sunspot. This model can then be used to classify previously unseen sunspots based on the candidate’s measurements. The generalization performance of a learned model (how closely the target outputs and the model’s predicted outputs agree for patterns that have not been presented to the learning algorithm) would provide an indication of how well the model has learned the desired mapping. More formally, a classification learning algorithm L takes a training set T as its input. The training set consists of |T| examples or instances. It is assumed that there is a probability distribution D from which all training examples are drawn independently—that is, all the training examples are independently and identically distributed (i.i.d.). The ith training example is of the form (x_i, y_i), where x_i is a vector of values of several features and y_i represents the class to be predicted.* In the sunspot classification example given above, each training example would represent one sunspot’s classification (y_i) and the corresponding set of measurements (x_i). The output of a supervised learning algorithm is a model h that approximates the unknown mapping from the inputs to the outputs. In our example, h would map from the sunspot measurements to the type of sunspot. We may have a test set S—a set of examples not used in training that we use to test how well the model h predicts the outputs on new examples. Just as with the examples in T, the examples in S are assumed to be independent and identically distributed (i.i.d.) draws from the distribution D. We measure the error of h on the test set as the proportion of test cases that h misclassifies: 1/|S| Sigma(x,y union S)[I(h(x)!= y)] where I(v) is the indicator function—it returns 1 if v is true and 0 otherwise. In our sunspot classification example, we would identify additional examples of sunspots that were not used in generating the model, and use these to determine how accurate the model is—the fraction of the test samples that the model classifies correctly. An example of a classification model is the decision tree shown in Figure 23.1. We will discuss the decision tree learning algorithm in more detail later—for now, we assume that, given a training set with examples of sunspots, this decision tree is derived. This can be used to classify previously unseen examples of sunpots. For example, if a new sunspot’s inputs indicate that its "Group Length" is in the range 10-15, then the decision tree would classify the sunspot as being of type “E,” whereas if the "Group Length" is "NULL," the "Magnetic Type" is "bipolar," and the "Penumbra" is "rudimentary," then it would be classified as type "C." In this chapter, we will add to the above description of classification problems. We will discuss decision trees and several other classification models. In particular, we will discuss the learning algorithms that generate these classification models, how to use them to classify new examples, and the strengths and weaknesses of these models. We will end with pointers to further reading on classification methods applied to astronomy data.

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics.

PubMed

Huang, Shujun; Cai, Nianguang; Pacheco, Pedro Penzuti; Narrandes, Shavira; Wang, Yang; Xu, Wayne

2018-01-01

Machine learning with maximization (support) of separating margin (vector), called support vector machine (SVM) learning, is a powerful classification tool that has been used for cancer genomic classification or subtyping. Today, as advancements in high-throughput technologies lead to production of large amounts of genomic and epigenomic data, the classification feature of SVMs is expanding its use in cancer genomics, leading to the discovery of new biomarkers, new drug targets, and a better understanding of cancer driver genes. Herein we reviewed the recent progress of SVMs in cancer genomic studies. We intend to comprehend the strength of the SVM learning and its future perspective in cancer genomic applications. Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.

Use of data mining techniques to classify soil CO2 emission induced by crop management in sugarcane field.

PubMed

Farhate, Camila Viana Vieira; Souza, Zigomar Menezes de; Oliveira, Stanley Robson de Medeiros; Tavares, Rose Luiza Moraes; Carvalho, João Luís Nunes

2018-01-01

Soil CO2 emissions are regarded as one of the largest flows of the global carbon cycle and small changes in their magnitude can have a large effect on the CO2 concentration in the atmosphere. Thus, a better understanding of this attribute would enable the identification of promoters and the development of strategies to mitigate the risks of climate change. Therefore, our study aimed at using data mining techniques to predict the soil CO2 emission induced by crop management in sugarcane areas in Brazil. To do so, we used different variable selection methods (correlation, chi-square, wrapper) and classification (Decision tree, Bayesian models, neural networks, support vector machine, bagging with logistic regression), and finally we tested the efficiency of different approaches through the Receiver Operating Characteristic (ROC) curve. The original dataset consisted of 19 variables (18 independent variables and one dependent (or response) variable). The association between cover crop and minimum tillage are effective strategies to promote the mitigation of soil CO2 emissions, in which the average CO2 emissions are 63 kg ha-1 day-1. The variables soil moisture, soil temperature (Ts), rainfall, pH, and organic carbon were most frequently selected for soil CO2 emission classification using different methods for attribute selection. According to the results of the ROC curve, the best approaches for soil CO2 emission classification were the following: (I)-the Multilayer Perceptron classifier with attribute selection through the wrapper method, that presented rate of false positive of 13,50%, true positive of 94,20% area under the curve (AUC) of 89,90% (II)-the Bagging classifier with logistic regression with attribute selection through the Chi-square method, that presented rate of false positive of 13,50%, true positive of 94,20% AUC of 89,90%. However, the (I) approach stands out in relation to (II) for its higher positive class accuracy (high CO2 emission) and lower computational cost.

Use of data mining techniques to classify soil CO2 emission induced by crop management in sugarcane field

PubMed Central

de Souza, Zigomar Menezes; Oliveira, Stanley Robson de Medeiros; Tavares, Rose Luiza Moraes; Carvalho, João Luís Nunes

2018-01-01

Soil CO2 emissions are regarded as one of the largest flows of the global carbon cycle and small changes in their magnitude can have a large effect on the CO2 concentration in the atmosphere. Thus, a better understanding of this attribute would enable the identification of promoters and the development of strategies to mitigate the risks of climate change. Therefore, our study aimed at using data mining techniques to predict the soil CO2 emission induced by crop management in sugarcane areas in Brazil. To do so, we used different variable selection methods (correlation, chi-square, wrapper) and classification (Decision tree, Bayesian models, neural networks, support vector machine, bagging with logistic regression), and finally we tested the efficiency of different approaches through the Receiver Operating Characteristic (ROC) curve. The original dataset consisted of 19 variables (18 independent variables and one dependent (or response) variable). The association between cover crop and minimum tillage are effective strategies to promote the mitigation of soil CO2 emissions, in which the average CO2 emissions are 63 kg ha-1 day-1. The variables soil moisture, soil temperature (Ts), rainfall, pH, and organic carbon were most frequently selected for soil CO2 emission classification using different methods for attribute selection. According to the results of the ROC curve, the best approaches for soil CO2 emission classification were the following: (I)–the Multilayer Perceptron classifier with attribute selection through the wrapper method, that presented rate of false positive of 13,50%, true positive of 94,20% area under the curve (AUC) of 89,90% (II)–the Bagging classifier with logistic regression with attribute selection through the Chi-square method, that presented rate of false positive of 13,50%, true positive of 94,20% AUC of 89,90%. However, the (I) approach stands out in relation to (II) for its higher positive class accuracy (high CO2 emission) and lower computational cost. PMID:29513765

Clay Stabilization Using the Ash of Mount Sinabung in Terms of the Value of California Bearing Ratio (CBR)

NASA Astrophysics Data System (ADS)

Hastuty, I. P.; Roesyanto, R.; Napitupulu, S. M. A.

2018-02-01

Most areas in Indonesia consist of clay soils with high plasticity so that to meet technical requirements the soil needs improvement, which is known as soil stabilization.There are three ways of soil stabilization process, i.e. mechanical, physical and chemical. In this study, chemical stabilization was performed, that was by adding stabilizing agents to the soil. The stabilizing agent used was the ash of Mount Sinabung. Since 2010 until now, Sinabung Mountain is still experiencing eruption that produces a lot of volcanic ash and it inconveniences the environment. So, it is expected that this research will be able to optimize the utilization of Sinabung ash. The purpose of this study was to investigate the effect of the addition of Mount Sinabung ash to CBR (California Bearing Ratio) value, to determine the effect of the curing time of one day and fourteen days mixture on the CBR value, and to find the mixed content with effective curing time to produce the largest CBR value. Based on this study, the soil type CL (Clay - Low Plasticity) was obtained, based on the classification of USCS (Unified Soil Classification System) and categorized as A-6 (6) based on the classification of AASHTO (American Association of State Highway and Transportation officials) with the most effective mixed stabilizer material which was the variation of 10% Mount Sinabung ash with fourteen days of curing time. The CBR value resulted from the mixture of 10% Sinabung ash that was cured within fourteen days was 8.95%. By the increase of the content of the Sinabung ash, the CBR value always improved to the level of 10%, Sinabung ash then decreased and became constant at the mixture of higher volcanic ash mixture but remained above the CBR value of the original soil.

A study of the effectiveness of the use of gypsum and volcanic ash against the stability of clay soil in terms of UCT and CBR values

NASA Astrophysics Data System (ADS)

Roesyanto; Iskandar, R.; Hastuty, IP; Lubis, AIU

2018-02-01

Soil stabilization is an effort to improve engineering properties of soil. The conventional soil stabilization is by adding additives to the soil such as Portland cement, lime, and bitumen. The clay stabilization research was done by adding gypsum and volcanic ash. The research purposes were to find out the value of engineering properties of clay due to the addition of 2% gypsum and 2% - 15% volcanic ash. The soil was classified as Clay - Low Plasticity (CL) based on USCS and was classified as A-7-6 (10) based on AASHTO classification system. The UCT values of original soil and original soil plus 2% gypsum were 1.40 kg/cm2 and 1.66 kg/cm2 respectively. The CBR soaked and unsoaked values of original soil were 4.44% and 6.28% correspondingly. Meanwhile, CBR soaked and CBR unsoaked values of original soil plus 2% gypsum were 6.74% and 8.02% respectively. The research results showed that the additives materials of gypsum and volcanic ash improved the engineering properties of clay. The UCT result from the stabilized soil by 2% gypsum and 10% volcanic ash gave value of 2.79 kg/cm2 (increased 99.28% from original soil). For CBR test, the most effective mixture were in variation of 2% gypsum and 9% volcanic ash which gave value of 9.07% (104.27% increase from original soil) for CBR soaked and 10.29% (63.85% increase from original soil) for CBR unsoaked. The stabilized soil with 2% gypsum and 9% volcanic ash was classified as CL based on USCS and was classified as A-6 (4) based on AASHTO classification system.

Global similarity predicts dissociation of classification and recognition: evidence questioning the implicit-explicit learning distinction in amnesia.

PubMed

Jamieson, Randall K; Holmes, Signy; Mewhort, D J K

2010-11-01

Dissociation of classification and recognition in amnesia is widely taken to imply 2 functional systems: an implicit procedural-learning system that is spared in amnesia and an explicit episodic-learning system that is compromised. We argue that both tasks reflect the global similarity of probes to memory. In classification, subjects sort unstudied grammatical exemplars from lures, whereas in recognition, they sort studied grammatical exemplars from lures. Hence, global similarity is necessarily greater in recognition than in classification. Moreover, a grammatical exemplar's similarity to studied exemplars is a nonlinear function of the integrity of the data in memory. Assuming that data integrity is better for control subjects than for subjects with amnesia, the nonlinear relation combined with the advantage for recognition over classification predicts the dissociation of recognition and classification. To illustrate the dissociation of recognition and classification in healthy undergraduates, we manipulated study time to vary the integrity of the data in memory and brought the dissociation under experimental control. We argue that the dissociation reflects a general cost in memory rather than a selective impairment of separate procedural and episodic systems. (c) 2010 APA, all rights reserved

A classification model of Hyperion image base on SAM combined decision tree

NASA Astrophysics Data System (ADS)

Wang, Zhenghai; Hu, Guangdao; Zhou, YongZhang; Liu, Xin

2009-10-01

Monitoring the Earth using imaging spectrometers has necessitated more accurate analyses and new applications to remote sensing. A very high dimensional input space requires an exponentially large amount of data to adequately and reliably represent the classes in that space. On the other hand, with increase in the input dimensionality the hypothesis space grows exponentially, which makes the classification performance highly unreliable. Traditional classification algorithms Classification of hyperspectral images is challenging. New algorithms have to be developed for hyperspectral data classification. The Spectral Angle Mapper (SAM) is a physically-based spectral classification that uses an ndimensional angle to match pixels to reference spectra. The algorithm determines the spectral similarity between two spectra by calculating the angle between the spectra, treating them as vectors in a space with dimensionality equal to the number of bands. The key and difficulty is that we should artificial defining the threshold of SAM. The classification precision depends on the rationality of the threshold of SAM. In order to resolve this problem, this paper proposes a new automatic classification model of remote sensing image using SAM combined with decision tree. It can automatic choose the appropriate threshold of SAM and improve the classify precision of SAM base on the analyze of field spectrum. The test area located in Heqing Yunnan was imaged by EO_1 Hyperion imaging spectrometer using 224 bands in visual and near infrared. The area included limestone areas, rock fields, soil and forests. The area was classified into four different vegetation and soil types. The results show that this method choose the appropriate threshold of SAM and eliminates the disturbance and influence of unwanted objects effectively, so as to improve the classification precision. Compared with the likelihood classification by field survey data, the classification precision of this model heightens 9.9%.

Compression of deep convolutional neural network for computer-aided diagnosis of masses in digital breast tomosynthesis

NASA Astrophysics Data System (ADS)

Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir; Helvie, Mark A.; Richter, Caleb; Cha, Kenny

2018-02-01

Deep-learning models are highly parameterized, causing difficulty in inference and transfer learning. We propose a layered pathway evolution method to compress a deep convolutional neural network (DCNN) for classification of masses in DBT while maintaining the classification accuracy. Two-stage transfer learning was used to adapt the ImageNet-trained DCNN to mammography and then to DBT. In the first-stage transfer learning, transfer learning from ImageNet trained DCNN was performed using mammography data. In the second-stage transfer learning, the mammography-trained DCNN was trained on the DBT data using feature extraction from fully connected layer, recursive feature elimination and random forest classification. The layered pathway evolution encapsulates the feature extraction to the classification stages to compress the DCNN. Genetic algorithm was used in an iterative approach with tournament selection driven by count-preserving crossover and mutation to identify the necessary nodes in each convolution layer while eliminating the redundant nodes. The DCNN was reduced by 99% in the number of parameters and 95% in mathematical operations in the convolutional layers. The lesion-based area under the receiver operating characteristic curve on an independent DBT test set from the original and the compressed network resulted in 0.88+/-0.05 and 0.90+/-0.04, respectively. The difference did not reach statistical significance. We demonstrated a DCNN compression approach without additional fine-tuning or loss of performance for classification of masses in DBT. The approach can be extended to other DCNNs and transfer learning tasks. An ensemble of these smaller and focused DCNNs has the potential to be used in multi-target transfer learning.

Machine learning methods for the classification of gliomas: Initial results using features extracted from MR spectroscopy.

PubMed

Ranjith, G; Parvathy, R; Vikas, V; Chandrasekharan, Kesavadas; Nair, Suresh

2015-04-01

With the advent of new imaging modalities, radiologists are faced with handling increasing volumes of data for diagnosis and treatment planning. The use of automated and intelligent systems is becoming essential in such a scenario. Machine learning, a branch of artificial intelligence, is increasingly being used in medical image analysis applications such as image segmentation, registration and computer-aided diagnosis and detection. Histopathological analysis is currently the gold standard for classification of brain tumors. The use of machine learning algorithms along with extraction of relevant features from magnetic resonance imaging (MRI) holds promise of replacing conventional invasive methods of tumor classification. The aim of the study is to classify gliomas into benign and malignant types using MRI data. Retrospective data from 28 patients who were diagnosed with glioma were used for the analysis. WHO Grade II (low-grade astrocytoma) was classified as benign while Grade III (anaplastic astrocytoma) and Grade IV (glioblastoma multiforme) were classified as malignant. Features were extracted from MR spectroscopy. The classification was done using four machine learning algorithms: multilayer perceptrons, support vector machine, random forest and locally weighted learning. Three of the four machine learning algorithms gave an area under ROC curve in excess of 0.80. Random forest gave the best performance in terms of AUC (0.911) while sensitivity was best for locally weighted learning (86.1%). The performance of different machine learning algorithms in the classification of gliomas is promising. An even better performance may be expected by integrating features extracted from other MR sequences. © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.

Active learning for clinical text classification: is it better than random sampling?

PubMed

Figueroa, Rosa L; Zeng-Treitler, Qing; Ngo, Long H; Goryachev, Sergey; Wiechmann, Eduardo P

2012-01-01

This study explores active learning algorithms as a way to reduce the requirements for large training sets in medical text classification tasks. Three existing active learning algorithms (distance-based (DIST), diversity-based (DIV), and a combination of both (CMB)) were used to classify text from five datasets. The performance of these algorithms was compared to that of passive learning on the five datasets. We then conducted a novel investigation of the interaction between dataset characteristics and the performance results. Classification accuracy and area under receiver operating characteristics (ROC) curves for each algorithm at different sample sizes were generated. The performance of active learning algorithms was compared with that of passive learning using a weighted mean of paired differences. To determine why the performance varies on different datasets, we measured the diversity and uncertainty of each dataset using relative entropy and correlated the results with the performance differences. The DIST and CMB algorithms performed better than passive learning. With a statistical significance level set at 0.05, DIST outperformed passive learning in all five datasets, while CMB was found to be better than passive learning in four datasets. We found strong correlations between the dataset diversity and the DIV performance, as well as the dataset uncertainty and the performance of the DIST algorithm. For medical text classification, appropriate active learning algorithms can yield performance comparable to that of passive learning with considerably smaller training sets. In particular, our results suggest that DIV performs better on data with higher diversity and DIST on data with lower uncertainty.

Active learning for clinical text classification: is it better than random sampling?

PubMed Central

Figueroa, Rosa L; Ngo, Long H; Goryachev, Sergey; Wiechmann, Eduardo P

2012-01-01

Objective This study explores active learning algorithms as a way to reduce the requirements for large training sets in medical text classification tasks. Design Three existing active learning algorithms (distance-based (DIST), diversity-based (DIV), and a combination of both (CMB)) were used to classify text from five datasets. The performance of these algorithms was compared to that of passive learning on the five datasets. We then conducted a novel investigation of the interaction between dataset characteristics and the performance results. Measurements Classification accuracy and area under receiver operating characteristics (ROC) curves for each algorithm at different sample sizes were generated. The performance of active learning algorithms was compared with that of passive learning using a weighted mean of paired differences. To determine why the performance varies on different datasets, we measured the diversity and uncertainty of each dataset using relative entropy and correlated the results with the performance differences. Results The DIST and CMB algorithms performed better than passive learning. With a statistical significance level set at 0.05, DIST outperformed passive learning in all five datasets, while CMB was found to be better than passive learning in four datasets. We found strong correlations between the dataset diversity and the DIV performance, as well as the dataset uncertainty and the performance of the DIST algorithm. Conclusion For medical text classification, appropriate active learning algorithms can yield performance comparable to that of passive learning with considerably smaller training sets. In particular, our results suggest that DIV performs better on data with higher diversity and DIST on data with lower uncertainty. PMID:22707743

SDL: Saliency-Based Dictionary Learning Framework for Image Similarity.

PubMed

Sarkar, Rituparna; Acton, Scott T

2018-02-01

In image classification, obtaining adequate data to learn a robust classifier has often proven to be difficult in several scenarios. Classification of histological tissue images for health care analysis is a notable application in this context due to the necessity of surgery, biopsy or autopsy. To adequately exploit limited training data in classification, we propose a saliency guided dictionary learning method and subsequently an image similarity technique for histo-pathological image classification. Salient object detection from images aids in the identification of discriminative image features. We leverage the saliency values for the local image regions to learn a dictionary and respective sparse codes for an image, such that the more salient features are reconstructed with smaller error. The dictionary learned from an image gives a compact representation of the image itself and is capable of representing images with similar content, with comparable sparse codes. We employ this idea to design a similarity measure between a pair of images, where local image features of one image, are encoded with the dictionary learned from the other and vice versa. To effectively utilize the learned dictionary, we take into account the contribution of each dictionary atom in the sparse codes to generate a global image representation for image comparison. The efficacy of the proposed method was evaluated using three tissue data sets that consist of mammalian kidney, lung and spleen tissue, breast cancer, and colon cancer tissue images. From the experiments, we observe that our methods outperform the state of the art with an increase of 14.2% in the average classification accuracy over all data sets.

Learning Styles among Students in an Advanced Soil Management Class: Impact on Students' Performance

ERIC Educational Resources Information Center

Eudoxie, Gaius D.

2011-01-01

Learning styles represent an integral component of the learning environment, which has been shown to differ across institutions and disciplines. To identify learner preferences within a discipline would aid in evaluating instructional resources geared toward active learning. The learning profiles of second-year soil science students (n = 62) were…

BCDForest: a boosting cascade deep forest model towards the classification of cancer subtypes based on gene expression data.

PubMed

Guo, Yang; Liu, Shuhui; Li, Zhanhuai; Shang, Xuequn

2018-04-11

The classification of cancer subtypes is of great importance to cancer disease diagnosis and therapy. Many supervised learning approaches have been applied to cancer subtype classification in the past few years, especially of deep learning based approaches. Recently, the deep forest model has been proposed as an alternative of deep neural networks to learn hyper-representations by using cascade ensemble decision trees. It has been proved that the deep forest model has competitive or even better performance than deep neural networks in some extent. However, the standard deep forest model may face overfitting and ensemble diversity challenges when dealing with small sample size and high-dimensional biology data. In this paper, we propose a deep learning model, so-called BCDForest, to address cancer subtype classification on small-scale biology datasets, which can be viewed as a modification of the standard deep forest model. The BCDForest distinguishes from the standard deep forest model with the following two main contributions: First, a named multi-class-grained scanning method is proposed to train multiple binary classifiers to encourage diversity of ensemble. Meanwhile, the fitting quality of each classifier is considered in representation learning. Second, we propose a boosting strategy to emphasize more important features in cascade forests, thus to propagate the benefits of discriminative features among cascade layers to improve the classification performance. Systematic comparison experiments on both microarray and RNA-Seq gene expression datasets demonstrate that our method consistently outperforms the state-of-the-art methods in application of cancer subtype classification. The multi-class-grained scanning and boosting strategy in our model provide an effective solution to ease the overfitting challenge and improve the robustness of deep forest model working on small-scale data. Our model provides a useful approach to the classification of cancer subtypes by using deep learning on high-dimensional and small-scale biology data.

Biological classification with RNA-Seq data: Can alternatively spliced transcript expression enhance machine learning classifier?

PubMed

Johnson, Nathan T; Dhroso, Andi; Hughes, Katelyn J; Korkin, Dmitry

2018-06-25

The extent to which the genes are expressed in the cell can be simplistically defined as a function of one or more factors of the environment, lifestyle, and genetics. RNA sequencing (RNA-Seq) is becoming a prevalent approach to quantify gene expression, and is expected to gain better insights to a number of biological and biomedical questions, compared to the DNA microarrays. Most importantly, RNA-Seq allows to quantify expression at the gene and alternative splicing isoform levels. However, leveraging the RNA-Seq data requires development of new data mining and analytics methods. Supervised machine learning methods are commonly used approaches for biological data analysis, and have recently gained attention for their applications to the RNA-Seq data. In this work, we assess the utility of supervised learning methods trained on RNA-Seq data for a diverse range of biological classification tasks. We hypothesize that the isoform-level expression data is more informative for biological classification tasks than the gene-level expression data. Our large-scale assessment is done through utilizing multiple datasets, organisms, lab groups, and RNA-Seq analysis pipelines. Overall, we performed and assessed 61 biological classification problems that leverage three independent RNA-Seq datasets and include over 2,000 samples that come from multiple organisms, lab groups, and RNA-Seq analyses. These 61 problems include predictions of the tissue type, sex, or age of the sample, healthy or cancerous phenotypes and, the pathological tumor stage for the samples from the cancerous tissue. For each classification problem, the performance of three normalization techniques and six machine learning classifiers was explored. We find that for every single classification problem, the isoform-based classifiers outperform or are comparable with gene expression based methods. The top-performing supervised learning techniques reached a near perfect classification accuracy, demonstrating the utility of supervised learning for RNA-Seq based data analysis. Published by Cold Spring Harbor Laboratory Press for the RNA Society.

An application to pulmonary emphysema classification based on model of texton learning by sparse representation

NASA Astrophysics Data System (ADS)

Zhang, Min; Zhou, Xiangrong; Goshima, Satoshi; Chen, Huayue; Muramatsu, Chisako; Hara, Takeshi; Yokoyama, Ryojiro; Kanematsu, Masayuki; Fujita, Hiroshi

2012-03-01

We aim at using a new texton based texture classification method in the classification of pulmonary emphysema in computed tomography (CT) images of the lungs. Different from conventional computer-aided diagnosis (CAD) pulmonary emphysema classification methods, in this paper, firstly, the dictionary of texton is learned via applying sparse representation(SR) to image patches in the training dataset. Then the SR coefficients of the test images over the dictionary are used to construct the histograms for texture presentations. Finally, classification is performed by using a nearest neighbor classifier with a histogram dissimilarity measure as distance. The proposed approach is tested on 3840 annotated regions of interest consisting of normal tissue and mild, moderate and severe pulmonary emphysema of three subtypes. The performance of the proposed system, with an accuracy of about 88%, is comparably higher than state of the art method based on the basic rotation invariant local binary pattern histograms and the texture classification method based on texton learning by k-means, which performs almost the best among other approaches in the literature.

Voice based gender classification using machine learning

NASA Astrophysics Data System (ADS)

Raahul, A.; Sapthagiri, R.; Pankaj, K.; Vijayarajan, V.

2017-11-01

Gender identification is one of the major problem speech analysis today. Tracing the gender from acoustic data i.e., pitch, median, frequency etc. Machine learning gives promising results for classification problem in all the research domains. There are several performance metrics to evaluate algorithms of an area. Our Comparative model algorithm for evaluating 5 different machine learning algorithms based on eight different metrics in gender classification from acoustic data. Agenda is to identify gender, with five different algorithms: Linear Discriminant Analysis (LDA), K-Nearest Neighbour (KNN), Classification and Regression Trees (CART), Random Forest (RF), and Support Vector Machine (SVM) on basis of eight different metrics. The main parameter in evaluating any algorithms is its performance. Misclassification rate must be less in classification problems, which says that the accuracy rate must be high. Location and gender of the person have become very crucial in economic markets in the form of AdSense. Here with this comparative model algorithm, we are trying to assess the different ML algorithms and find the best fit for gender classification of acoustic data.

Development and Application of a Soil Moisture Downscaling Method for Mobility Assessment

DTIC Science & Technology

2011-05-01

instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send...REPORT Development and Application of a Soil Moisture Downscaling Method for Mobility Assessment 14. ABSTRACT 16. SECURITY CLASSIFICATION OF: Soil...cells). Thus, a method is required to downscale intermediate-resolution patterns to finer resolutions. Fortunately, fine-resolution variations in

Do We Need a New Definition of Soil?

NASA Astrophysics Data System (ADS)

Arnold, Richard W.; Brevik, Eric C.

2014-05-01

Effective communication is really desirable to better relate with politicians, an interested lay public, and others not involved in soil science. Soil survey programs are intended to help people understand how soils function in their landscapes to make ecosystems operate better without damaging the environment and to indicate different kinds of suitability for various purposes. The properties of soils as recognized, described, and mapped at detailed scales form the basis for developing diagnostics for a systematic taxonomy that enables scientists to interact with other. In the USA mapping done at scales of 1:15,840± made it possible to define and use so-called "soil series", initially as soil map units, but later as central concepts of a set of soils which could be segregated using phases to indicate important features, primarily for farming. Detailed soil surveys published using a standard format helps maintain uniformity across the country. Soil series are recognized as the basic units of soils within the evolving hierarchical soil taxonomy and diagnostic properties are defined, measured and used to update and modify the scientific classification. Concepts like soil quality and soil function are considered to be "attributes" and not basic properties of soils. They are the collective interpretation of the combination of properties thought to be relevant for communicating important aspects of using, managing, restoring, and protecting the lands of any locality, region, or country. A famous example in the US was the land capability system with classes and subclasses of suitability for agricultural land uses. An updated soil survey in California contains over 500 pages providing details about classes of 30 different functional soil classifications for 155 map units. Over the years soil extension agents were the interpreters of the science to the lay folks and could help them form mental pictures of soils and soil landscapes locally They were the early leaders of what we think of as "field guides to natural resources" such as trees, flowers, birds, and so forth. There were not such books to identify soils but the basics have always been there waiting for proper attention, preparation, and use. At smaller scales the map units are always combinations of the basic units, and now it is possible to use some higher category classes to indicate the central concepts of larger areas. Every year soil scientists around the world observe and describe features and properties of soils in landscapes that are getting more attention than previously. Soil genesis studies help us to better understand the complexity of landscape and soil evolution. Often they indicate that current soils are commonly being formed from parts of previous soils. We do not need a new definition of soil. We do need to work on developing and testing complete interpretive classifications of soils to better meet the needs of societies today. This means "soil quality", "soil functions", and other attributes of soils require more attention, now and in the near future to permit politicians and lay publics to better understand the significance of soils to the future of civilization. "After all is said and done, more is said than done" Aesop, Greek storyteller

A statistical approach for validating eSOTER and digital soil maps in front of traditional soil maps

NASA Astrophysics Data System (ADS)

Bock, Michael; Baritz, Rainer; Köthe, Rüdiger; Melms, Stephan; Günther, Susann

2015-04-01

During the European research project eSOTER, three different Digital Soil Maps (DSM) were developed for the pilot area Chemnitz 1:250,000 (FP7 eSOTER project, grant agreement nr. 211578). The core task of the project was to revise the SOTER method for the interpretation of soil and terrain data. It was one of the working hypothesis that eSOTER does not only provide terrain data with typical soil profiles, but that the new products actually perform like a conceptual soil map. The three eSOTER maps for the pilot area considerably differed in spatial representation and content of soil classes. In this study we compare the three eSOTER maps against existing reconnaissance soil maps keeping in mind that traditional soil maps have many subjective issues and intended bias regarding the overestimation and emphasize of certain features. Hence, a true validation of the proper representation of modeled soil maps is hardly possible; rather a statistical comparison between modeled and empirical approaches is possible. If eSOTER data represent conceptual soil maps, then different eSOTER, DSM and conventional maps from various sources and different regions could be harmonized towards consistent new data sets for large areas including the whole European continent. One of the eSOTER maps has been developed closely to the traditional SOTER method: terrain classification data (derived from SRTM DEM) were combined with lithology data (re-interpreted geological map); the corresponding terrain units were then extended with soil information: a very dense regional soil profile data set was used to define soil mapping units based on a statistical grouping of terrain units. The second map is a pure DSM map using continuous terrain parameters instead of terrain classification; radiospectrometric data were used to supplement parent material information from geology maps. The classification method Random Forest was used. The third approach predicts soil diagnostic properties based on covariates similar to DSM practices; in addition, multi-temporal MODIS data were used; the resulting soil map is the product of these diagnostic layers producing a map of soil reference groups (classified according to WRB). Because the third approach was applied to a larger test area in central Europe, and compared to the first two approaches, has worked with coarser input data, comparability is only partly fulfilled. To evaluate the usability of the three eSOTER maps, and to make a comparison among them, traditional soil maps 1:200,000 and 1:50,000 were used as reference data sets. Three statistical methods were applied: (i) in a moving window the distribution of the soil classes of each DSM product was compared to that of the soil maps by calculating the corrected coefficient of contingency, (ii) the value of predictive power for each of the eSOTER maps was determined, and (iii) the degree of consistency was derived. The latter is based on a weighting of the match of occurring class combinations via expert knowledge and recalculating the proportions of map appearance with these weights. To re-check the validation results a field study by local soil experts was conducted. The results show clearly that the first eSOTER approach based on the terrain classification / reinterpreted parent material information has the greatest similarity with traditional soil maps. The spatial differentiation offered by such an approach is well suitable to serve as a conceptual soil map. Therefore, eSOTER can be a tool for soil mappers to generate conceptual soil maps in a faster and more consistent way. This conclusion is at least valid for overview scales such as 1.250,000.

Couple Graph Based Label Propagation Method for Hyperspectral Remote Sensing Data Classification

NASA Astrophysics Data System (ADS)

Wang, X. P.; Hu, Y.; Chen, J.

2018-04-01

Graph based semi-supervised classification method are widely used for hyperspectral image classification. We present a couple graph based label propagation method, which contains both the adjacency graph and the similar graph. We propose to construct the similar graph by using the similar probability, which utilize the label similarity among examples probably. The adjacency graph was utilized by a common manifold learning method, which has effective improve the classification accuracy of hyperspectral data. The experiments indicate that the couple graph Laplacian which unite both the adjacency graph and the similar graph, produce superior classification results than other manifold Learning based graph Laplacian and Sparse representation based graph Laplacian in label propagation framework.

Alzheimer's disease detection via automatic 3D caudate nucleus segmentation using coupled dictionary learning with level set formulation.

PubMed

Al-Shaikhli, Saif Dawood Salman; Yang, Michael Ying; Rosenhahn, Bodo

2016-12-01

This paper presents a novel method for Alzheimer's disease classification via an automatic 3D caudate nucleus segmentation. The proposed method consists of segmentation and classification steps. In the segmentation step, we propose a novel level set cost function. The proposed cost function is constrained by a sparse representation of local image features using a dictionary learning method. We present coupled dictionaries: a feature dictionary of a grayscale brain image and a label dictionary of a caudate nucleus label image. Using online dictionary learning, the coupled dictionaries are learned from the training data. The learned coupled dictionaries are embedded into a level set function. In the classification step, a region-based feature dictionary is built. The region-based feature dictionary is learned from shape features of the caudate nucleus in the training data. The classification is based on the measure of the similarity between the sparse representation of region-based shape features of the segmented caudate in the test image and the region-based feature dictionary. The experimental results demonstrate the superiority of our method over the state-of-the-art methods by achieving a high segmentation (91.5%) and classification (92.5%) accuracy. In this paper, we find that the study of the caudate nucleus atrophy gives an advantage over the study of whole brain structure atrophy to detect Alzheimer's disease. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Predominant-period site classification for response spectra prediction equations in Italy

USGS Publications Warehouse

Di Alessandro, Carola; Bonilla, Luis Fabian; Boore, David M.; Rovelli, Antonio; Scotti, Oona

2012-01-01

We propose a site‐classification scheme based on the predominant period of the site, as determined from the average horizontal‐to‐vertical (H/V) spectral ratios of ground motion. Our scheme extends Zhao et al. (2006) classifications by adding two classes, the most important of which is defined by flat H/V ratios with amplitudes less than 2. The proposed classification is investigated by using 5%‐damped response spectra from Italian earthquake records. We select a dataset of 602 three‐component analog and digital recordings from 120 earthquakes recorded at 214 seismic stations within a hypocentral distance of 200 km. Selected events are in the moment‐magnitude range 4.0≤Mw≤6.8 and focal depths from a few kilometers to 46 km. We computed H/V ratios for these data and used them to classify each site into one of six classes. We then investigate the impact of this classification scheme on empirical ground‐motion prediction equations (GMPEs) by comparing its performance with that of the conventional rock/soil classification. Although the adopted approach results in only a small reduction of the overall standard deviation, the use of H/V spectral ratios in site classification does capture the signature of sites with flat frequency‐response, as well as deep and shallow‐soil profiles, characterized by long‐ and short‐period resonance, respectively; in addition, the classification scheme is relatively quick and inexpensive, which is an advantage over schemes based on measurements of shear‐wave velocity.

Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning

NASA Technical Reports Server (NTRS)

Fayyad, U.; Irani, K.

1993-01-01

Since most real-world applications of classification learning involve continuous-valued attributes, properly addressing the discretization process is an important problem. This paper addresses the use of the entropy minimization heuristic for discretizing the range of a continuous-valued attribute into multiple intervals.

Deep learning decision fusion for the classification of urban remote sensing data

NASA Astrophysics Data System (ADS)

Abdi, Ghasem; Samadzadegan, Farhad; Reinartz, Peter

2018-01-01

Multisensor data fusion is one of the most common and popular remote sensing data classification topics by considering a robust and complete description about the objects of interest. Furthermore, deep feature extraction has recently attracted significant interest and has become a hot research topic in the geoscience and remote sensing research community. A deep learning decision fusion approach is presented to perform multisensor urban remote sensing data classification. After deep features are extracted by utilizing joint spectral-spatial information, a soft-decision made classifier is applied to train high-level feature representations and to fine-tune the deep learning framework. Next, a decision-level fusion classifies objects of interest by the joint use of sensors. Finally, a context-aware object-based postprocessing is used to enhance the classification results. A series of comparative experiments are conducted on the widely used dataset of 2014 IEEE GRSS data fusion contest. The obtained results illustrate the considerable advantages of the proposed deep learning decision fusion over the traditional classifiers.

The Iterated Classification Game: A New Model of the Cultural Transmission of Language

PubMed Central

Swarup, Samarth; Gasser, Les

2010-01-01

The Iterated Classification Game (ICG) combines the Classification Game with the Iterated Learning Model (ILM) to create a more realistic model of the cultural transmission of language through generations. It includes both learning from parents and learning from peers. Further, it eliminates some of the chief criticisms of the ILM: that it does not study grounded languages, that it does not include peer learning, and that it builds in a bias for compositional languages. We show that, over the span of a few generations, a stable linguistic system emerges that can be acquired very quickly by each generation, is compositional, and helps the agents to solve the classification problem with which they are faced. The ICG also leads to a different interpretation of the language acquisition process. It suggests that the role of parents is to initialize the linguistic system of the child in such a way that subsequent interaction with peers results in rapid convergence to the correct language. PMID:20190877

Comparing statistical and machine learning classifiers: alternatives for predictive modeling in human factors research.

PubMed

Carnahan, Brian; Meyer, Gérard; Kuntz, Lois-Ann

2003-01-01

Multivariate classification models play an increasingly important role in human factors research. In the past, these models have been based primarily on discriminant analysis and logistic regression. Models developed from machine learning research offer the human factors professional a viable alternative to these traditional statistical classification methods. To illustrate this point, two machine learning approaches--genetic programming and decision tree induction--were used to construct classification models designed to predict whether or not a student truck driver would pass his or her commercial driver license (CDL) examination. The models were developed and validated using the curriculum scores and CDL exam performances of 37 student truck drivers who had completed a 320-hr driver training course. Results indicated that the machine learning classification models were superior to discriminant analysis and logistic regression in terms of predictive accuracy. Actual or potential applications of this research include the creation of models that more accurately predict human performance outcomes.

Discriminative Nonlinear Analysis Operator Learning: When Cosparse Model Meets Image Classification.

PubMed

Wen, Zaidao; Hou, Biao; Jiao, Licheng

2017-05-03

Linear synthesis model based dictionary learning framework has achieved remarkable performances in image classification in the last decade. Behaved as a generative feature model, it however suffers from some intrinsic deficiencies. In this paper, we propose a novel parametric nonlinear analysis cosparse model (NACM) with which a unique feature vector will be much more efficiently extracted. Additionally, we derive a deep insight to demonstrate that NACM is capable of simultaneously learning the task adapted feature transformation and regularization to encode our preferences, domain prior knowledge and task oriented supervised information into the features. The proposed NACM is devoted to the classification task as a discriminative feature model and yield a novel discriminative nonlinear analysis operator learning framework (DNAOL). The theoretical analysis and experimental performances clearly demonstrate that DNAOL will not only achieve the better or at least competitive classification accuracies than the state-of-the-art algorithms but it can also dramatically reduce the time complexities in both training and testing phases.

An incremental approach to genetic-algorithms-based classification.

PubMed

Guan, Sheng-Uei; Zhu, Fangming

2005-04-01

Incremental learning has been widely addressed in the machine learning literature to cope with learning tasks where the learning environment is ever changing or training samples become available over time. However, most research work explores incremental learning with statistical algorithms or neural networks, rather than evolutionary algorithms. The work in this paper employs genetic algorithms (GAs) as basic learning algorithms for incremental learning within one or more classifier agents in a multiagent environment. Four new approaches with different initialization schemes are proposed. They keep the old solutions and use an "integration" operation to integrate them with new elements to accommodate new attributes, while biased mutation and crossover operations are adopted to further evolve a reinforced solution. The simulation results on benchmark classification data sets show that the proposed approaches can deal with the arrival of new input attributes and integrate them with the original input space. It is also shown that the proposed approaches can be successfully used for incremental learning and improve classification rates as compared to the retraining GA. Possible applications for continuous incremental training and feature selection are also discussed.

Machine learning vortices at the Kosterlitz-Thouless transition

NASA Astrophysics Data System (ADS)

Beach, Matthew J. S.; Golubeva, Anna; Melko, Roger G.

2018-01-01

Efficient and automated classification of phases from minimally processed data is one goal of machine learning in condensed-matter and statistical physics. Supervised algorithms trained on raw samples of microstates can successfully detect conventional phase transitions via learning a bulk feature such as an order parameter. In this paper, we investigate whether neural networks can learn to classify phases based on topological defects. We address this question on the two-dimensional classical XY model which exhibits a Kosterlitz-Thouless transition. We find significant feature engineering of the raw spin states is required to convincingly claim that features of the vortex configurations are responsible for learning the transition temperature. We further show a single-layer network does not correctly classify the phases of the XY model, while a convolutional network easily performs classification by learning the global magnetization. Finally, we design a deep network capable of learning vortices without feature engineering. We demonstrate the detection of vortices does not necessarily result in the best classification accuracy, especially for lattices of less than approximately 1000 spins. For larger systems, it remains a difficult task to learn vortices.

Test Operation Procedure (TOP) 01-1-010A Vehicle Test Course Severity (Surface Roughness)

DTIC Science & Technology

2017-12-12

Department of Agriculture (USDA) classifications, respectively. TABLE 10. PARTICLE SIZE CLASSES CLASS SIZE Cobble and Gravel >4.75 mm particle diameter...ABBREVIATIONS. USCS Unified Soil Classification System USDA United States Department of Agriculture UTM Universal Transverse Mercator WNS wave number

Classification of spatially unresolved objects

NASA Technical Reports Server (NTRS)

Nalepka, R. F.; Horwitz, H. M.; Hyde, P. D.; Morgenstern, J. P.

1972-01-01

A proportion estimation technique for classification of multispectral scanner images is reported that uses data point averaging to extract and compute estimated proportions for a single average data point to classify spatial unresolved areas. Example extraction calculations of spectral signatures for bare soil, weeds, alfalfa, and barley prove quite accurate.

Application of the SNoW machine learning paradigm to a set of transportation imaging problems

NASA Astrophysics Data System (ADS)

Paul, Peter; Burry, Aaron M.; Wang, Yuheng; Kozitsky, Vladimir

2012-01-01

Machine learning methods have been successfully applied to image object classification problems where there is clear distinction between classes and where a comprehensive set of training samples and ground truth are readily available. The transportation domain is an area where machine learning methods are particularly applicable, since the classification problems typically have well defined class boundaries and, due to high traffic volumes in most applications, massive roadway data is available. Though these classes tend to be well defined, the particular image noises and variations can be challenging. Another challenge is the extremely high accuracy typically required in most traffic applications. Incorrect assignment of fines or tolls due to imaging mistakes is not acceptable in most applications. For the front seat vehicle occupancy detection problem, classification amounts to determining whether one face (driver only) or two faces (driver + passenger) are detected in the front seat of a vehicle on a roadway. For automatic license plate recognition, the classification problem is a type of optical character recognition problem encompassing multiple class classification. The SNoW machine learning classifier using local SMQT features is shown to be successful in these two transportation imaging applications.

A novel approach to malignant-benign classification of pulmonary nodules by using ensemble learning classifiers.

PubMed

Tartar, A; Akan, A; Kilic, N

2014-01-01

Computer-aided detection systems can help radiologists to detect pulmonary nodules at an early stage. In this paper, a novel Computer-Aided Diagnosis system (CAD) is proposed for the classification of pulmonary nodules as malignant and benign. The proposed CAD system using ensemble learning classifiers, provides an important support to radiologists at the diagnosis process of the disease, achieves high classification performance. The proposed approach with bagging classifier results in 94.7 %, 90.0 % and 77.8 % classification sensitivities for benign, malignant and undetermined classes (89.5 % accuracy), respectively.

Hydrogeomorphic Classification of Wetlands on Mt. Desert Island, Maine, Including Hydrologic Susceptibility Factors for Wetlands in Acadia National Park

USGS Publications Warehouse

Nielsen, Martha G.

2006-01-01

The U.S. Geological Survey, in cooperation with the National Park Service, developed a hydrogeomorphic (HGM) classification system for wetlands greater than 0.4 hectares (ha) on Mt. Desert Island, Maine, and applied this classification using map-scale data to more than 1,200 mapped wetland units on the island. In addition, two hydrologic susceptibility factors were defined for a subset of these wetlands, using 11 variables derived from landscape-scale characteristics of the catchment areas of these wetlands. The hydrologic susceptibility factors, one related to the potential hydrologic pathways for contaminants and the other to the susceptibility of wetlands to disruptions in water supply from projected future changes in climate, were used to indicate which wetlands (greater than 1 ha) in Acadia National Park (ANP) may warrant further investigation or monitoring. The HGM classification system consists of 13 categories: Riverine-Upper Perennial, Riverine-Nonperennial, Riverine- Tidal, Depressional-Closed, Depressional-Semiclosed, Depressional-Open, Depressional-No Ground-Water Input, Mineral Soil Flat, Organic Soil Flat, Tidal Fringe, Lacustrine Fringe, Slope, and Hilltop/Upper Hillslope. A dichotomous key was developed to aid in the classification of wetlands. The National Wetland Inventory maps produced by the U.S. Fish and Wildlife Service provided the wetland mapping units used for this classification. On the basis of topographic map information and geographic information system (GIS) layers at a scale of 1:24,000 or larger, 1,202 wetland units were assigned a preliminary HGM classification. Two of the 13 HGM classes (Riverine-Tidal and Depressional-No Ground-Water Input) were not assigned to any wetlands because criteria for determining those classes are not available at that map scale, and must be determined by more site-specific information. Of the 1,202 wetland polygons classified, which cover 1,830 ha in ANP, 327 were classified as Slope, 258 were Depressional (Open, Semiclosed, and Closed), 231 were Riverine (Upper Perennial and Nonperennial), 210 were Soil Flat (Mineral and Organic), 68 were Lacustrine Fringe, 51 were Tidal Fringe, 22 were Hilltop/Upper Hillslope, and another 35 were small open water bodies. Most small, isolated wetlands classified on the island are Slope wetlands. The least common, Hilltop/Upper Hillslope wetlands, only occur on a few hilltops and shoulders of hills and mountains. Large wetland complexes generally consist of groups of Depressional wetlands and Mineral Soil Flat or Organic Soil Flat wetlands, often with fringing Slope wetlands at their edges and Riverine wetlands near streams flowing through them. The two analyses of wetland hydrologic susceptibility on Mt. Desert Island were applied to 186 wetlands located partially or entirely within ANP. These analyses were conducted using individually mapped catchments for each wetland. The 186 wetlands were aggregated from the original 1,202 mapped wetland polygons on the basis of their HGM classes. Landscape-level hydrologic, geomorphic, and soil variables were defined for the catchments of the wetlands, and transformed into scaled scores from 0 to 10 for each variable. The variables included area of the wetland, area of the catchment, area of the wetland divided by the area of the catchment, the average topographic slope of the catchment, the amount of the catchment where bedrock crops out with no soil cover or excessively thin soil cover, the amount of storage (in lakes and wetlands) in the catchment, the topographic relief of the catchment, the amount of clay-rich soil in the catchment, the amount of manmade impervious surface, whether the wetland had a stream inflow, and whether the wetland had a hydraulic connection to a lake or estuary. These data were determined using a GIS and data layers mapped at a scale of 1:24,000 or larger. These landscape variables were combined in different ways for the two hydrologic susceptibility fact

[Severity classification of chronic obstructive pulmonary disease based on deep learning].

PubMed

Ying, Jun; Yang, Ceyuan; Li, Quanzheng; Xue, Wanguo; Li, Tanshi; Cao, Wenzhe

2017-12-01

In this paper, a deep learning method has been raised to build an automatic classification algorithm of severity of chronic obstructive pulmonary disease. Large sample clinical data as input feature were analyzed for their weights in classification. Through feature selection, model training, parameter optimization and model testing, a classification prediction model based on deep belief network was built to predict severity classification criteria raised by the Global Initiative for Chronic Obstructive Lung Disease (GOLD). We get accuracy over 90% in prediction for two different standardized versions of severity criteria raised in 2007 and 2011 respectively. Moreover, we also got the contribution ranking of different input features through analyzing the model coefficient matrix and confirmed that there was a certain degree of agreement between the more contributive input features and the clinical diagnostic knowledge. The validity of the deep belief network model was proved by this result. This study provides an effective solution for the application of deep learning method in automatic diagnostic decision making.

Convolutional neural network with transfer learning for rice type classification

NASA Astrophysics Data System (ADS)

Patel, Vaibhav Amit; Joshi, Manjunath V.

2018-04-01

Presently, rice type is identified manually by humans, which is time consuming and error prone. Therefore, there is a need to do this by machine which makes it faster with greater accuracy. This paper proposes a deep learning based method for classification of rice types. We propose two methods to classify the rice types. In the first method, we train a deep convolutional neural network (CNN) using the given segmented rice images. In the second method, we train a combination of a pretrained VGG16 network and the proposed method, while using transfer learning in which the weights of a pretrained network are used to achieve better accuracy. Our approach can also be used for classification of rice grain as broken or fine. We train a 5-class model for classifying rice types using 4000 training images and another 2- class model for the classification of broken and normal rice using 1600 training images. We observe that despite having distinct rice images, our architecture, pretrained on ImageNet data boosts classification accuracy significantly.

Can a Smartphone Diagnose Parkinson Disease? A Deep Neural Network Method and Telediagnosis System Implementation.

PubMed

Zhang, Y N

2017-01-01

Parkinson's disease (PD) is primarily diagnosed by clinical examinations, such as walking test, handwriting test, and MRI diagnostic. In this paper, we propose a machine learning based PD telediagnosis method for smartphone. Classification of PD using speech records is a challenging task owing to the fact that the classification accuracy is still lower than doctor-level. Here we demonstrate automatic classification of PD using time frequency features, stacked autoencoders (SAE), and K nearest neighbor (KNN) classifier. KNN classifier can produce promising classification results from useful representations which were learned by SAE. Empirical results show that the proposed method achieves better performance with all tested cases across classification tasks, demonstrating machine learning capable of classifying PD with a level of competence comparable to doctor. It concludes that a smartphone can therefore potentially provide low-cost PD diagnostic care. This paper also gives an implementation on browser/server system and reports the running time cost. Both advantages and disadvantages of the proposed telediagnosis system are discussed.

Can a Smartphone Diagnose Parkinson Disease? A Deep Neural Network Method and Telediagnosis System Implementation

PubMed Central

2017-01-01

Parkinson's disease (PD) is primarily diagnosed by clinical examinations, such as walking test, handwriting test, and MRI diagnostic. In this paper, we propose a machine learning based PD telediagnosis method for smartphone. Classification of PD using speech records is a challenging task owing to the fact that the classification accuracy is still lower than doctor-level. Here we demonstrate automatic classification of PD using time frequency features, stacked autoencoders (SAE), and K nearest neighbor (KNN) classifier. KNN classifier can produce promising classification results from useful representations which were learned by SAE. Empirical results show that the proposed method achieves better performance with all tested cases across classification tasks, demonstrating machine learning capable of classifying PD with a level of competence comparable to doctor. It concludes that a smartphone can therefore potentially provide low-cost PD diagnostic care. This paper also gives an implementation on browser/server system and reports the running time cost. Both advantages and disadvantages of the proposed telediagnosis system are discussed. PMID:29075547

A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update

NASA Astrophysics Data System (ADS)

Lotte, F.; Bougrain, L.; Cichocki, A.; Clerc, M.; Congedo, M.; Rakotomamonjy, A.; Yger, F.

2018-06-01

Objective. Most current electroencephalography (EEG)-based brain–computer interfaces (BCIs) are based on machine learning algorithms. There is a large diversity of classifier types that are used in this field, as described in our 2007 review paper. Now, approximately ten years after this review publication, many new algorithms have been developed and tested to classify EEG signals in BCIs. The time is therefore ripe for an updated review of EEG classification algorithms for BCIs. Approach. We surveyed the BCI and machine learning literature from 2007 to 2017 to identify the new classification approaches that have been investigated to design BCIs. We synthesize these studies in order to present such algorithms, to report how they were used for BCIs, what were the outcomes, and to identify their pros and cons. Main results. We found that the recently designed classification algorithms for EEG-based BCIs can be divided into four main categories: adaptive classifiers, matrix and tensor classifiers, transfer learning and deep learning, plus a few other miscellaneous classifiers. Among these, adaptive classifiers were demonstrated to be generally superior to static ones, even with unsupervised adaptation. Transfer learning can also prove useful although the benefits of transfer learning remain unpredictable. Riemannian geometry-based methods have reached state-of-the-art performances on multiple BCI problems and deserve to be explored more thoroughly, along with tensor-based methods. Shrinkage linear discriminant analysis and random forests also appear particularly useful for small training samples settings. On the other hand, deep learning methods have not yet shown convincing improvement over state-of-the-art BCI methods. Significance. This paper provides a comprehensive overview of the modern classification algorithms used in EEG-based BCIs, presents the principles of these methods and guidelines on when and how to use them. It also identifies a number of challenges to further advance EEG classification in BCI.

Dynamic Response of Reinforced Soil Systems. Volume 1. Report

DTIC Science & Technology

1993-03-01

include Security Clas~sification) DYNAMIC RWSPC!SE OF REIFý1Cý SOIL SYSTEM~, VCTJI4E I OF II: PREPO~r 󈧐. PERSONAL AUTHOR($) BMW3U, R.C.; FRAWASZY...protected by a burster slab. These protection measures are costly, time consuming to construct, and sensitive to multiple strikes. Soil has been used to...characterize the static load-deflection behavior of the reinforced soil. Dynamic pullout tests were then performed using the same parameters as the static

A study of the utilization of ERTS-1 data from the Wabash River Basin

NASA Technical Reports Server (NTRS)

Landgrebe, D. A. (Principal Investigator)

1973-01-01

The author has identified the following significant results. Nine projects are defined, five ERTS data applications experiments and four supporting technology tasks. The most significant applications results were achieved in the soil association mapping, earth surface feature identification, and urban land use mapping efforts. Four soil association boundaries were accurately delineated from ERTS-1 imagery. A data bank has been developed to test surface feature classifications obtained from ERTS-1 data. Preliminary forest cover classifications indicated that the number of acres estimated tended to be greater than actually existed by 25%. Urban land use analysis of ERTS-1 data indicated highly accurate classification could be obtained for many urban catagories. The wooded residential category tended to be misclassified as woods or agricultural land. Further statistical analysis revealed that these classes could be separated using sample variance.

Towards the development of multifunctional molecular indicators combining soil biogeochemical and microbiological variables to predict the ecological integrity of silvicultural practices.

PubMed

Peck, Vincent; Quiza, Liliana; Buffet, Jean-Philippe; Khdhiri, Mondher; Durand, Audrey-Anne; Paquette, Alain; Thiffault, Nelson; Messier, Christian; Beaulieu, Nadyre; Guertin, Claude; Constant, Philippe

2016-05-01

The impact of mechanical site preparation (MSP) on soil biogeochemical structure in young larch plantations was investigated. Soil samples were collected in replicated plots comprising simple trenching, double trenching, mounding and inverting site preparation. Unlogged natural mixed forest areas were used as a reference. Analysis of soil nutrients, abundance of bacteria and gas exchanges unveiled no significant difference among the plots. However, inverting site preparation resulted in higher variations of gas exchanges when compared with trenching, mounding and unlogged natural forest. A combination of the biological and physicochemical variables was used to define a multifunctional classification of the soil samples into four distinct groups categorized as a function of their deviation from baseline ecological conditions. According to this classification model, simple trenching was the approach that represented the lowest ecological risk potential at the microsite level. No relationship was observed between MSP method and soil bacterial community structure as assessed by high-throughput sequencing of bacterial 16S rRNA gene; however, indicator genotypes were identified for each multifunctional soil class. This is the first identification of multifunctional molecular indicators for baseline and disturbed ecological conditions in soil, demonstrating the potential of applied microbial ecology to guide silvicultural practices and ecological risk assessment. © 2016 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology.

ASIST SIG/CR Classification Workshop 2000: Classification for User Support and Learning.

ERIC Educational Resources Information Center

Soergel, Dagobert

2001-01-01

Reports on papers presented at the 62nd Annual Meeting of ASIST (American Society for Information Science and Technology) for the Special Interest Group in Classification Research (SIG/CR). Topics include types of knowledge; developing user-oriented classifications, including domain analysis; classification in the user interface; and automatic…

Naïve and Robust: Class-Conditional Independence in Human Classification Learning

ERIC Educational Resources Information Center

Jarecki, Jana B.; Meder, Björn; Nelson, Jonathan D.

2018-01-01

Humans excel in categorization. Yet from a computational standpoint, learning a novel probabilistic classification task involves severe computational challenges. The present paper investigates one way to address these challenges: assuming class-conditional independence of features. This feature independence assumption simplifies the inference…

A novel deep learning approach for classification of EEG motor imagery signals.

PubMed

Tabar, Yousef Rezaei; Halici, Ugur

2017-02-01

Signal classification is an important issue in brain computer interface (BCI) systems. Deep learning approaches have been used successfully in many recent studies to learn features and classify different types of data. However, the number of studies that employ these approaches on BCI applications is very limited. In this study we aim to use deep learning methods to improve classification performance of EEG motor imagery signals. In this study we investigate convolutional neural networks (CNN) and stacked autoencoders (SAE) to classify EEG Motor Imagery signals. A new form of input is introduced to combine time, frequency and location information extracted from EEG signal and it is used in CNN having one 1D convolutional and one max-pooling layers. We also proposed a new deep network by combining CNN and SAE. In this network, the features that are extracted in CNN are classified through the deep network SAE. The classification performance obtained by the proposed method on BCI competition IV dataset 2b in terms of kappa value is 0.547. Our approach yields 9% improvement over the winner algorithm of the competition. Our results show that deep learning methods provide better classification performance compared to other state of art approaches. These methods can be applied successfully to BCI systems where the amount of data is large due to daily recording.

Soil Management Plan for the Oak Ridge Y-12 National Security Complex Oak Ridge, Tennessee

DOE Office of Scientific and Technical Information (OSTI.GOV)

None

2005-03-02

This Soil Management Plan applies to all activities conducted under the auspices of the National Nuclear Security Administration (NNSA) Oak Ridge Y-12 National Security Complex (Y-12) that involve soil disturbance and potential management of waste soil. The plan was prepared under the direction of the Y-12 Environmental Compliance Department of the Environment, Safety, and Health Division. Soil disturbances related to maintenance activities, utility and building construction projects, or demolition projects fall within the purview of the plan. This Soil Management Plan represents an integrated, visually oriented, planning and information resource tool for decision making involving excavation or disturbance of soilmore » at Y-12. This Soil Management Plan addresses three primary elements. (1) Regulatory and programmatic requirements for management of soil based on the location of a soil disturbance project and/or the regulatory classification of any contaminants that may be present (Chap. 2). Five general regulatory or programmatic classifications of soil are recognized to be potentially present at Y-12; soil may fall under one or more these classifications: (a) Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) pursuant to the Oak Ridge Reservation (ORR) Federal Facilities Agreement; (b) Resource Conservation and Recovery Act (RCRA); (c) RCRA 3004(u) solid waste managements units pursuant to the RCRA Hazardous and Solid Waste Amendments Act of 1984 permit for the ORR; (d) Toxic Substances and Control Act-regulated soil containing polychlorinated biphenyls; and (e) Radiologically contaminated soil regulated under the Atomic Energy Act review process. (2) Information for project planners on current and future planned remedial actions (RAs), as prescribed by CERCLA decision documents (including the scope of the actions and remedial goals), land use controls implemented to support or maintain RAs, RCRA post-closure regulatory requirements for former waste management units, legacy contamination source areas and distribution of contamination in soils, and environmental infrastructure (e.g., caps, monitoring systems, etc.) that is in place or planned in association with RAs. (3) Regulatory considerations and processes for management and disposition of waste soil upon generation, including regulatory drivers, best management practices (BMPs), waste determination protocols, waste acceptance criteria, and existing waste management procedures and BMPs for Y-12. This Soil Management Plan provides information to project planners to better coordinate their activities with other organizations and programs with a vested interest in soil disturbance activities at Y-12. The information allows project managers and maintenance personnel to evaluate and anticipate potential contaminant levels that may be present at a proposed soil disturbance site prior to commencement of activities and allows a more accurate assessment of potential waste management requirements.« less

Deep learning for hybrid EEG-fNIRS brain-computer interface: application to motor imagery classification.

PubMed

Chiarelli, Antonio Maria; Croce, Pierpaolo; Merla, Arcangelo; Zappasodi, Filippo

2018-06-01

Brain-computer interface (BCI) refers to procedures that link the central nervous system to a device. BCI was historically performed using electroencephalography (EEG). In the last years, encouraging results were obtained by combining EEG with other neuroimaging technologies, such as functional near infrared spectroscopy (fNIRS). A crucial step of BCI is brain state classification from recorded signal features. Deep artificial neural networks (DNNs) recently reached unprecedented complex classification outcomes. These performances were achieved through increased computational power, efficient learning algorithms, valuable activation functions, and restricted or back-fed neurons connections. By expecting significant overall BCI performances, we investigated the capabilities of combining EEG and fNIRS recordings with state-of-the-art deep learning procedures. We performed a guided left and right hand motor imagery task on 15 subjects with a fixed classification response time of 1 s and overall experiment length of 10 min. Left versus right classification accuracy of a DNN in the multi-modal recording modality was estimated and it was compared to standalone EEG and fNIRS and other classifiers. At a group level we obtained significant increase in performance when considering multi-modal recordings and DNN classifier with synergistic effect. BCI performances can be significantly improved by employing multi-modal recordings that provide electrical and hemodynamic brain activity information, in combination with advanced non-linear deep learning classification procedures.

Deep learning for hybrid EEG-fNIRS brain–computer interface: application to motor imagery classification

NASA Astrophysics Data System (ADS)

Chiarelli, Antonio Maria; Croce, Pierpaolo; Merla, Arcangelo; Zappasodi, Filippo

2018-06-01

Objective. Brain–computer interface (BCI) refers to procedures that link the central nervous system to a device. BCI was historically performed using electroencephalography (EEG). In the last years, encouraging results were obtained by combining EEG with other neuroimaging technologies, such as functional near infrared spectroscopy (fNIRS). A crucial step of BCI is brain state classification from recorded signal features. Deep artificial neural networks (DNNs) recently reached unprecedented complex classification outcomes. These performances were achieved through increased computational power, efficient learning algorithms, valuable activation functions, and restricted or back-fed neurons connections. By expecting significant overall BCI performances, we investigated the capabilities of combining EEG and fNIRS recordings with state-of-the-art deep learning procedures. Approach. We performed a guided left and right hand motor imagery task on 15 subjects with a fixed classification response time of 1 s and overall experiment length of 10 min. Left versus right classification accuracy of a DNN in the multi-modal recording modality was estimated and it was compared to standalone EEG and fNIRS and other classifiers. Main results. At a group level we obtained significant increase in performance when considering multi-modal recordings and DNN classifier with synergistic effect. Significance. BCI performances can be significantly improved by employing multi-modal recordings that provide electrical and hemodynamic brain activity information, in combination with advanced non-linear deep learning classification procedures.

MARTA GANs: Unsupervised Representation Learning for Remote Sensing Image Classification

NASA Astrophysics Data System (ADS)

Lin, Daoyu; Fu, Kun; Wang, Yang; Xu, Guangluan; Sun, Xian

2017-11-01

With the development of deep learning, supervised learning has frequently been adopted to classify remotely sensed images using convolutional networks (CNNs). However, due to the limited amount of labeled data available, supervised learning is often difficult to carry out. Therefore, we proposed an unsupervised model called multiple-layer feature-matching generative adversarial networks (MARTA GANs) to learn a representation using only unlabeled data. MARTA GANs consists of both a generative model $G$ and a discriminative model $D$. We treat $D$ as a feature extractor. To fit the complex properties of remote sensing data, we use a fusion layer to merge the mid-level and global features. $G$ can produce numerous images that are similar to the training data; therefore, $D$ can learn better representations of remotely sensed images using the training data provided by $G$. The classification results on two widely used remote sensing image databases show that the proposed method significantly improves the classification performance compared with other state-of-the-art methods.

Integrated Low-Rank-Based Discriminative Feature Learning for Recognition.

PubMed

Zhou, Pan; Lin, Zhouchen; Zhang, Chao

2016-05-01

Feature learning plays a central role in pattern recognition. In recent years, many representation-based feature learning methods have been proposed and have achieved great success in many applications. However, these methods perform feature learning and subsequent classification in two separate steps, which may not be optimal for recognition tasks. In this paper, we present a supervised low-rank-based approach for learning discriminative features. By integrating latent low-rank representation (LatLRR) with a ridge regression-based classifier, our approach combines feature learning with classification, so that the regulated classification error is minimized. In this way, the extracted features are more discriminative for the recognition tasks. Our approach benefits from a recent discovery on the closed-form solutions to noiseless LatLRR. When there is noise, a robust Principal Component Analysis (PCA)-based denoising step can be added as preprocessing. When the scale of a problem is large, we utilize a fast randomized algorithm to speed up the computation of robust PCA. Extensive experimental results demonstrate the effectiveness and robustness of our method.

Coast redwood ecological types of southern Monterey County, California

Treesearch

Mark Borchert; Daniel Segotta; Michael D. Purser

1988-01-01

An ecological classification system has been developed for the Pacific Southwest Region of the Forest Service. As part of this classification effort, coast redwood (Sequoia sempervirens) forests of southern Monterey County in the Los Padres National Forest were classified into six ecological types using vegetation, soils and geomorphology taken from...

A Novel Extreme Learning Machine Classification Model for e-Nose Application Based on the Multiple Kernel Approach

PubMed Central

Jian, Yulin; Huang, Daoyu; Yan, Jia; Lu, Kun; Huang, Ying; Wen, Tailai; Zeng, Tanyue; Zhong, Shijie; Xie, Qilong

2017-01-01

A novel classification model, named the quantum-behaved particle swarm optimization (QPSO)-based weighted multiple kernel extreme learning machine (QWMK-ELM), is proposed in this paper. Experimental validation is carried out with two different electronic nose (e-nose) datasets. Being different from the existing multiple kernel extreme learning machine (MK-ELM) algorithms, the combination coefficients of base kernels are regarded as external parameters of single-hidden layer feedforward neural networks (SLFNs). The combination coefficients of base kernels, the model parameters of each base kernel, and the regularization parameter are optimized by QPSO simultaneously before implementing the kernel extreme learning machine (KELM) with the composite kernel function. Four types of common single kernel functions (Gaussian kernel, polynomial kernel, sigmoid kernel, and wavelet kernel) are utilized to constitute different composite kernel functions. Moreover, the method is also compared with other existing classification methods: extreme learning machine (ELM), kernel extreme learning machine (KELM), k-nearest neighbors (KNN), support vector machine (SVM), multi-layer perceptron (MLP), radical basis function neural network (RBFNN), and probabilistic neural network (PNN). The results have demonstrated that the proposed QWMK-ELM outperforms the aforementioned methods, not only in precision, but also in efficiency for gas classification. PMID:28629202

Accurate classification of brain gliomas by discriminate dictionary learning based on projective dictionary pair learning of proton magnetic resonance spectra.

PubMed

Adebileje, Sikiru Afolabi; Ghasemi, Keyvan; Aiyelabegan, Hammed Tanimowo; Saligheh Rad, Hamidreza

2017-04-01

Proton magnetic resonance spectroscopy is a powerful noninvasive technique that complements the structural images of cMRI, which aids biomedical and clinical researches, by identifying and visualizing the compositions of various metabolites within the tissues of interest. However, accurate classification of proton magnetic resonance spectroscopy is still a challenging issue in clinics due to low signal-to-noise ratio, overlapping peaks of metabolites, and the presence of background macromolecules. This paper evaluates the performance of a discriminate dictionary learning classifiers based on projective dictionary pair learning method for brain gliomas proton magnetic resonance spectroscopy spectra classification task, and the result were compared with the sub-dictionary learning methods. The proton magnetic resonance spectroscopy data contain a total of 150 spectra (74 healthy, 23 grade II, 23 grade III, and 30 grade IV) from two databases. The datasets from both databases were first coupled together, followed by column normalization. The Kennard-Stone algorithm was used to split the datasets into its training and test sets. Performance comparison based on the overall accuracy, sensitivity, specificity, and precision was conducted. Based on the overall accuracy of our classification scheme, the dictionary pair learning method was found to outperform the sub-dictionary learning methods 97.78% compared with 68.89%, respectively. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

A European Humus Forms Reference Base

NASA Astrophysics Data System (ADS)

Zanella, A.; Englisch, M.; Ponge, J.-F.; Jabiol, B.; Sartori, G.; Gardi, C.

2012-04-01

From 2003 on, a panel of experts in humus and humus dynamics (Humus group) has been working about a standardisation and improvement of existing national humus classifications. Some important goals have been reached, in order to share data and experiences: a) definition of specific terms; b) description of 15 types of diagnostic horizons; c) of 10 basic humus forms references; d) subdivision of each main reference in 2-4 sub-unities; e) elaboration of a general European Humus Form Reference Base (http://hal-agroparistech.archives-ouvertes.fr/docs/00/56/17/95/PDF/Humus_Forms_ERB_31_01_2011.pdf); f) publication of the scientific significance of this base of classification as an article [A European morpho-functional classification of humus forms. Geoderma, 164 (3-4), 138-145]. The classification will be updated every 2 years and presently the Humus group is assessing biological (general: soil, vegetation, biome; specific: fungi, bacteria, pedofauna), physical (air temperature, rainfall) and chemical (pH, mineral elements, organic matter, quality and quantity of humic components…) factors which characterize basic humus forms and their varieties. The content of the new version of the classification is planned to be more "practical", like an ecological manual which lists associated humus forms and environmental data in the aim to contribute to a more precise environmental diagnosis of every analysed terrestrial and semiterrestrial European ecosystem. The Humus group is also involved in an endeavour to include humus forms in the World Reference Base for Soils (WRB-FAO) according to nomenclatural principles erected for soil profiles. Thirty basic references have been defined, complemented by a set of qualifiers (prefixes and suffixes), allowing to classify European humus forms and probably a large majority of humus forms known worldwide. The principles of the classification, the diagnostic horizons and humus forms main references are presented at the General Assembly of the European Geosciences Union with the aim to stimulate members' curiosity. Interested people are invited to test the classification system in various field areas and to collaborate with the Humus group. Critical observations and field data/impressions are welcome as every other suggestions which can help in elaborating the 2013 version of the European humus forms classification.

Instructional Method Classifications Lack User Language and Orientation

ERIC Educational Resources Information Center

Neumann, Susanne; Koper, Rob

2010-01-01

Following publications emphasizing the need of a taxonomy for instructional methods, this article presents a literature review on classifications for learning and teaching in order to identify possible classifications for instructional methods. Data was collected for 37 classifications capturing the origins, theoretical underpinnings, purposes and…

Classification

NASA Technical Reports Server (NTRS)

Oza, Nikunj C.

2011-01-01

A supervised learning task involves constructing a mapping from input data (normally described by several features) to the appropriate outputs. Within supervised learning, one type of task is a classification learning task, in which each output is one or more classes to which the input belongs. In supervised learning, a set of training examples---examples with known output values---is used by a learning algorithm to generate a model. This model is intended to approximate the mapping between the inputs and outputs. This model can be used to generate predicted outputs for inputs that have not been seen before. For example, we may have data consisting of observations of sunspots. In a classification learning task, our goal may be to learn to classify sunspots into one of several types. Each example may correspond to one candidate sunspot with various measurements or just an image. A learning algorithm would use the supplied examples to generate a model that approximates the mapping between each supplied set of measurements and the type of sunspot. This model can then be used to classify previously unseen sunspots based on the candidate's measurements. This chapter discusses methods to perform machine learning, with examples involving astronomy.

Multi-channel EEG-based sleep stage classification with joint collaborative representation and multiple kernel learning.

PubMed

Shi, Jun; Liu, Xiao; Li, Yan; Zhang, Qi; Li, Yingjie; Ying, Shihui

2015-10-30

Electroencephalography (EEG) based sleep staging is commonly used in clinical routine. Feature extraction and representation plays a crucial role in EEG-based automatic classification of sleep stages. Sparse representation (SR) is a state-of-the-art unsupervised feature learning method suitable for EEG feature representation. Collaborative representation (CR) is an effective data coding method used as a classifier. Here we use CR as a data representation method to learn features from the EEG signal. A joint collaboration model is established to develop a multi-view learning algorithm, and generate joint CR (JCR) codes to fuse and represent multi-channel EEG signals. A two-stage multi-view learning-based sleep staging framework is then constructed, in which JCR and joint sparse representation (JSR) algorithms first fuse and learning the feature representation from multi-channel EEG signals, respectively. Multi-view JCR and JSR features are then integrated and sleep stages recognized by a multiple kernel extreme learning machine (MK-ELM) algorithm with grid search. The proposed two-stage multi-view learning algorithm achieves superior performance for sleep staging. With a K-means clustering based dictionary, the mean classification accuracy, sensitivity and specificity are 81.10 ± 0.15%, 71.42 ± 0.66% and 94.57 ± 0.07%, respectively; while with the dictionary learned using the submodular optimization method, they are 80.29 ± 0.22%, 71.26 ± 0.78% and 94.38 ± 0.10%, respectively. The two-stage multi-view learning based sleep staging framework outperforms all other classification methods compared in this work, while JCR is superior to JSR. The proposed multi-view learning framework has the potential for sleep staging based on multi-channel or multi-modality polysomnography signals. Copyright © 2015 Elsevier B.V. All rights reserved.

Rapid Erosion Modeling in a Western Kenya Watershed using Visible Near Infrared Reflectance, Classification Tree Analysis and 137Cesium.

PubMed

deGraffenried, Jeff B; Shepherd, Keith D

2009-12-15

Human induced soil erosion has severe economic and environmental impacts throughout the world. It is more severe in the tropics than elsewhere and results in diminished food production and security. Kenya has limited arable land and 30 percent of the country experiences severe to very severe human induced soil degradation. The purpose of this research was to test visible near infrared diffuse reflectance spectroscopy (VNIR) as a tool for rapid assessment and benchmarking of soil condition and erosion severity class. The study was conducted in the Saiwa River watershed in the northern Rift Valley Province of western Kenya, a tropical highland area. Soil 137 Cs concentration was measured to validate spectrally derived erosion classes and establish the background levels for difference land use types. Results indicate VNIR could be used to accurately evaluate a large and diverse soil data set and predict soil erosion characteristics. Soil condition was spectrally assessed and modeled. Analysis of mean raw spectra indicated significant reflectance differences between soil erosion classes. The largest differences occurred between 1,350 and 1,950 nm with the largest separation occurring at 1,920 nm. Classification and Regression Tree (CART) analysis indicated that the spectral model had practical predictive success (72%) with Receiver Operating Characteristic (ROC) of 0.74. The change in 137 Cs concentrations supported the premise that VNIR is an effective tool for rapid screening of soil erosion condition.

Variation in soil macro-fauna diversity in seven humus orders of a Parrotio-Carpinetum forest association on Chromic Cambisols of Shast-klateh area in Iran

NASA Astrophysics Data System (ADS)

Izadi, M.; Habashi, H.; Waez-Mousavi, S. M.

2017-03-01

Soil biodiversity includes organisms which spend a part or all of their life cycle on or in the soil. Among soil-dwelling animals, macro-fauna as an important group of animals have important effects on the dynamics of soil organic matter and litter decomposition process. The humus forms interact with the climatic conditions, flora, as well as soil fauna, and microbial activity. In new humus form classifications, soil organisms play an important role in separation of humus horizons from one another. The subject of this study was to determine the diversity of macro fauna for different humus forms. We determined humus forms using morphological classification, and then 69 random samples were taken from plots of 100 cm2 in area, and soil macro-fauna species were collected by hand sorting method. Two classes of humus forms, including Mull (with three humus orders, namely Dysmull, Oligomull, and Mesomull,) and Amphi (with four humus orders, namely Leptoamphi, Eumacroamphi, Eumesoamphi, and Pachyamphi) were identified. A number of 13 macro-fauna orders were identified using identification key. Among the humus orders, Shannon diversity, Simpson evenness and Margalef richness indices were the highest in Pachyamphi order. Arthropod diversity in Pachyamphi humus order was higher than those of Mull. These results showed that diversity of soil macrofauna increase by increasing the thickness of the organic horizons (OL, OF, OH), especially OH horizon.

Teaching/Learning Methods and Students' Classification of Food Items

ERIC Educational Resources Information Center

Hamilton-Ekeke, Joy-Telu; Thomas, Malcolm

2011-01-01

Purpose: This study aims to investigate the effectiveness of a teaching method (TLS (Teaching/Learning Sequence)) based on a social constructivist paradigm on students' conceptualisation of classification of food. Design/methodology/approach: The study compared the TLS model developed by the researcher based on the social constructivist paradigm…

Obtaining Accurate Probabilities Using Classifier Calibration

ERIC Educational Resources Information Center

Pakdaman Naeini, Mahdi

2016-01-01

Learning probabilistic classification and prediction models that generate accurate probabilities is essential in many prediction and decision-making tasks in machine learning and data mining. One way to achieve this goal is to post-process the output of classification models to obtain more accurate probabilities. These post-processing methods are…

Transfer Learning beyond Text Classification

NASA Astrophysics Data System (ADS)

Yang, Qiang

Transfer learning is a new machine learning and data mining framework that allows the training and test data to come from different distributions or feature spaces. We can find many novel applications of machine learning and data mining where transfer learning is necessary. While much has been done in transfer learning in text classification and reinforcement learning, there has been a lack of documented success stories of novel applications of transfer learning in other areas. In this invited article, I will argue that transfer learning is in fact quite ubiquitous in many real world applications. In this article, I will illustrate this point through an overview of a broad spectrum of applications of transfer learning that range from collaborative filtering to sensor based location estimation and logical action model learning for AI planning. I will also discuss some potential future directions of transfer learning.

EMG finger movement classification based on ANFIS

NASA Astrophysics Data System (ADS)

Caesarendra, W.; Tjahjowidodo, T.; Nico, Y.; Wahyudati, S.; Nurhasanah, L.

2018-04-01

An increase number of people suffering from stroke has impact to the rapid development of finger hand exoskeleton to enable an automatic physical therapy. Prior to the development of finger exoskeleton, a research topic yet important i.e. machine learning of finger gestures classification is conducted. This paper presents a study on EMG signal classification of 5 finger gestures as a preliminary study toward the finger exoskeleton design and development in Indonesia. The EMG signals of 5 finger gestures were acquired using Myo EMG sensor. The EMG signal features were extracted and reduced using PCA. The ANFIS based learning is used to classify reduced features of 5 finger gestures. The result shows that the classification of finger gestures is less than the classification of 7 hand gestures.

Choosing the Most Effective Pattern Classification Model under Learning-Time Constraint.

PubMed

Saito, Priscila T M; Nakamura, Rodrigo Y M; Amorim, Willian P; Papa, João P; de Rezende, Pedro J; Falcão, Alexandre X

2015-01-01

Nowadays, large datasets are common and demand faster and more effective pattern analysis techniques. However, methodologies to compare classifiers usually do not take into account the learning-time constraints required by applications. This work presents a methodology to compare classifiers with respect to their ability to learn from classification errors on a large learning set, within a given time limit. Faster techniques may acquire more training samples, but only when they are more effective will they achieve higher performance on unseen testing sets. We demonstrate this result using several techniques, multiple datasets, and typical learning-time limits required by applications.

Deep Learning in Label-free Cell Classification

PubMed Central

Chen, Claire Lifan; Mahjoubfar, Ata; Tai, Li-Chia; Blaby, Ian K.; Huang, Allen; Niazi, Kayvan Reza; Jalali, Bahram

2016-01-01

Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individual cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. This system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells. PMID:26975219

Emotional textile image classification based on cross-domain convolutional sparse autoencoders with feature selection

NASA Astrophysics Data System (ADS)

Li, Zuhe; Fan, Yangyu; Liu, Weihua; Yu, Zeqi; Wang, Fengqin

2017-01-01

We aim to apply sparse autoencoder-based unsupervised feature learning to emotional semantic analysis for textile images. To tackle the problem of limited training data, we present a cross-domain feature learning scheme for emotional textile image classification using convolutional autoencoders. We further propose a correlation-analysis-based feature selection method for the weights learned by sparse autoencoders to reduce the number of features extracted from large size images. First, we randomly collect image patches on an unlabeled image dataset in the source domain and learn local features with a sparse autoencoder. We then conduct feature selection according to the correlation between different weight vectors corresponding to the autoencoder's hidden units. We finally adopt a convolutional neural network including a pooling layer to obtain global feature activations of textile images in the target domain and send these global feature vectors into logistic regression models for emotional image classification. The cross-domain unsupervised feature learning method achieves 65% to 78% average accuracy in the cross-validation experiments corresponding to eight emotional categories and performs better than conventional methods. Feature selection can reduce the computational cost of global feature extraction by about 50% while improving classification performance.

Deep Learning in Label-free Cell Classification

NASA Astrophysics Data System (ADS)

Chen, Claire Lifan; Mahjoubfar, Ata; Tai, Li-Chia; Blaby, Ian K.; Huang, Allen; Niazi, Kayvan Reza; Jalali, Bahram

2016-03-01

Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individual cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. This system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.

An Evaluation of Feature Learning Methods for High Resolution Image Classification

NASA Astrophysics Data System (ADS)

Tokarczyk, P.; Montoya, J.; Schindler, K.

2012-07-01

Automatic image classification is one of the fundamental problems of remote sensing research. The classification problem is even more challenging in high-resolution images of urban areas, where the objects are small and heterogeneous. Two questions arise, namely which features to extract from the raw sensor data to capture the local radiometry and image structure at each pixel or segment, and which classification method to apply to the feature vectors. While classifiers are nowadays well understood, selecting the right features remains a largely empirical process. Here we concentrate on the features. Several methods are evaluated which allow one to learn suitable features from unlabelled image data by analysing the image statistics. In a comparative study, we evaluate unsupervised feature learning with different linear and non-linear learning methods, including principal component analysis (PCA) and deep belief networks (DBN). We also compare these automatically learned features with popular choices of ad-hoc features including raw intensity values, standard combinations like the NDVI, a few PCA channels, and texture filters. The comparison is done in a unified framework using the same images, the target classes, reference data and a Random Forest classifier.

Pulmonary emphysema classification based on an improved texton learning model by sparse representation

NASA Astrophysics Data System (ADS)

Zhang, Min; Zhou, Xiangrong; Goshima, Satoshi; Chen, Huayue; Muramatsu, Chisako; Hara, Takeshi; Yokoyama, Ryujiro; Kanematsu, Masayuki; Fujita, Hiroshi

2013-03-01

In this paper, we present a texture classification method based on texton learned via sparse representation (SR) with new feature histogram maps in the classification of emphysema. First, an overcomplete dictionary of textons is learned via KSVD learning on every class image patches in the training dataset. In this stage, high-pass filter is introduced to exclude patches in smooth area to speed up the dictionary learning process. Second, 3D joint-SR coefficients and intensity histograms of the test images are used for characterizing regions of interest (ROIs) instead of conventional feature histograms constructed from SR coefficients of the test images over the dictionary. Classification is then performed using a classifier with distance as a histogram dissimilarity measure. Four hundreds and seventy annotated ROIs extracted from 14 test subjects, including 6 paraseptal emphysema (PSE) subjects, 5 centrilobular emphysema (CLE) subjects and 3 panlobular emphysema (PLE) subjects, are used to evaluate the effectiveness and robustness of the proposed method. The proposed method is tested on 167 PSE, 240 CLE and 63 PLE ROIs consisting of mild, moderate and severe pulmonary emphysema. The accuracy of the proposed system is around 74%, 88% and 89% for PSE, CLE and PLE, respectively.

Deep learning for brain tumor classification

NASA Astrophysics Data System (ADS)

Paul, Justin S.; Plassard, Andrew J.; Landman, Bennett A.; Fabbri, Daniel

2017-03-01

Recent research has shown that deep learning methods have performed well on supervised machine learning, image classification tasks. The purpose of this study is to apply deep learning methods to classify brain images with different tumor types: meningioma, glioma, and pituitary. A dataset was publicly released containing 3,064 T1-weighted contrast enhanced MRI (CE-MRI) brain images from 233 patients with either meningioma, glioma, or pituitary tumors split across axial, coronal, or sagittal planes. This research focuses on the 989 axial images from 191 patients in order to avoid confusing the neural networks with three different planes containing the same diagnosis. Two types of neural networks were used in classification: fully connected and convolutional neural networks. Within these two categories, further tests were computed via the augmentation of the original 512×512 axial images. Training neural networks over the axial data has proven to be accurate in its classifications with an average five-fold cross validation of 91.43% on the best trained neural network. This result demonstrates that a more general method (i.e. deep learning) can outperform specialized methods that require image dilation and ring-forming subregions on tumors.

Discriminative Dictionary Learning With Two-Level Low Rank and Group Sparse Decomposition for Image Classification.

PubMed

Wen, Zaidao; Hou, Zaidao; Jiao, Licheng

2017-11-01

Discriminative dictionary learning (DDL) framework has been widely used in image classification which aims to learn some class-specific feature vectors as well as a representative dictionary according to a set of labeled training samples. However, interclass similarities and intraclass variances among input samples and learned features will generally weaken the representability of dictionary and the discrimination of feature vectors so as to degrade the classification performance. Therefore, how to explicitly represent them becomes an important issue. In this paper, we present a novel DDL framework with two-level low rank and group sparse decomposition model. In the first level, we learn a class-shared and several class-specific dictionaries, where a low rank and a group sparse regularization are, respectively, imposed on the corresponding feature matrices. In the second level, the class-specific feature matrix will be further decomposed into a low rank and a sparse matrix so that intraclass variances can be separated to concentrate the corresponding feature vectors. Extensive experimental results demonstrate the effectiveness of our model. Compared with the other state-of-the-arts on several popular image databases, our model can achieve a competitive or better performance in terms of the classification accuracy.

Biomarkers for Musculoskeletal Pain Conditions: Use of Brain Imaging and Machine Learning.

PubMed

Boissoneault, Jeff; Sevel, Landrew; Letzen, Janelle; Robinson, Michael; Staud, Roland

2017-01-01

Chronic musculoskeletal pain condition often shows poor correlations between tissue abnormalities and clinical pain. Therefore, classification of pain conditions like chronic low back pain, osteoarthritis, and fibromyalgia depends mostly on self report and less on objective findings like X-ray or magnetic resonance imaging (MRI) changes. However, recent advances in structural and functional brain imaging have identified brain abnormalities in chronic pain conditions that can be used for illness classification. Because the analysis of complex and multivariate brain imaging data is challenging, machine learning techniques have been increasingly utilized for this purpose. The goal of machine learning is to train specific classifiers to best identify variables of interest on brain MRIs (i.e., biomarkers). This report describes classification techniques capable of separating MRI-based brain biomarkers of chronic pain patients from healthy controls with high accuracy (70-92%) using machine learning, as well as critical scientific, practical, and ethical considerations related to their potential clinical application. Although self-report remains the gold standard for pain assessment, machine learning may aid in the classification of chronic pain disorders like chronic back pain and fibromyalgia as well as provide mechanistic information regarding their neural correlates.

Fines classification based on sensitivity to pore-fluid chemistry

USGS Publications Warehouse

Jang, Junbong; Santamarina, J. Carlos

2016-01-01

The 75-μm particle size is used to discriminate between fine and coarse grains. Further analysis of fine grains is typically based on the plasticity chart. Whereas pore-fluid-chemistry-dependent soil response is a salient and distinguishing characteristic of fine grains, pore-fluid chemistry is not addressed in current classification systems. Liquid limits obtained with electrically contrasting pore fluids (deionized water, 2-M NaCl brine, and kerosene) are combined to define the soil “electrical sensitivity.” Liquid limit and electrical sensitivity can be effectively used to classify fine grains according to their fluid-soil response into no-, low-, intermediate-, or high-plasticity fine grains of low, intermediate, or high electrical sensitivity. The proposed methodology benefits from the accumulated experience with liquid limit in the field and addresses the needs of a broader range of geotechnical engineering problems.

Using machine learning techniques to automate sky survey catalog generation

NASA Technical Reports Server (NTRS)

Fayyad, Usama M.; Roden, J. C.; Doyle, R. J.; Weir, Nicholas; Djorgovski, S. G.

1993-01-01

We describe the application of machine classification techniques to the development of an automated tool for the reduction of a large scientific data set. The 2nd Palomar Observatory Sky Survey provides comprehensive photographic coverage of the northern celestial hemisphere. The photographic plates are being digitized into images containing on the order of 10(exp 7) galaxies and 10(exp 8) stars. Since the size of this data set precludes manual analysis and classification of objects, our approach is to develop a software system which integrates independently developed techniques for image processing and data classification. Image processing routines are applied to identify and measure features of sky objects. Selected features are used to determine the classification of each object. GID3* and O-BTree, two inductive learning techniques, are used to automatically learn classification decision trees from examples. We describe the techniques used, the details of our specific application, and the initial encouraging results which indicate that our approach is well-suited to the problem. The benefits of the approach are increased data reduction throughput, consistency of classification, and the automated derivation of classification rules that will form an objective, examinable basis for classifying sky objects. Furthermore, astronomers will be freed from the tedium of an intensely visual task to pursue more challenging analysis and interpretation problems given automatically cataloged data.

A Robust Deep Model for Improved Classification of AD/MCI Patients

PubMed Central

Li, Feng; Tran, Loc; Thung, Kim-Han; Ji, Shuiwang; Shen, Dinggang; Li, Jiang

2015-01-01

Accurate classification of Alzheimer’s Disease (AD) and its prodromal stage, Mild Cognitive Impairment (MCI), plays a critical role in possibly preventing progression of memory impairment and improving quality of life for AD patients. Among many research tasks, it is of particular interest to identify noninvasive imaging biomarkers for AD diagnosis. In this paper, we present a robust deep learning system to identify different progression stages of AD patients based on MRI and PET scans. We utilized the dropout technique to improve classical deep learning by preventing its weight co-adaptation, which is a typical cause of over-fitting in deep learning. In addition, we incorporated stability selection, an adaptive learning factor, and a multi-task learning strategy into the deep learning framework. We applied the proposed method to the ADNI data set and conducted experiments for AD and MCI conversion diagnosis. Experimental results showed that the dropout technique is very effective in AD diagnosis, improving the classification accuracies by 5.9% on average as compared to the classical deep learning methods. PMID:25955998

Active learning methods for interactive image retrieval.

PubMed

Gosselin, Philippe Henri; Cord, Matthieu

2008-07-01

Active learning methods have been considered with increased interest in the statistical learning community. Initially developed within a classification framework, a lot of extensions are now being proposed to handle multimedia applications. This paper provides algorithms within a statistical framework to extend active learning for online content-based image retrieval (CBIR). The classification framework is presented with experiments to compare several powerful classification techniques in this information retrieval context. Focusing on interactive methods, active learning strategy is then described. The limitations of this approach for CBIR are emphasized before presenting our new active selection process RETIN. First, as any active method is sensitive to the boundary estimation between classes, the RETIN strategy carries out a boundary correction to make the retrieval process more robust. Second, the criterion of generalization error to optimize the active learning selection is modified to better represent the CBIR objective of database ranking. Third, a batch processing of images is proposed. Our strategy leads to a fast and efficient active learning scheme to retrieve sets of online images (query concept). Experiments on large databases show that the RETIN method performs well in comparison to several other active strategies.

Coupled dimensionality reduction and classification for supervised and semi-supervised multilabel learning

PubMed Central

Gönen, Mehmet

2014-01-01

Coupled training of dimensionality reduction and classification is proposed previously to improve the prediction performance for single-label problems. Following this line of research, in this paper, we first introduce a novel Bayesian method that combines linear dimensionality reduction with linear binary classification for supervised multilabel learning and present a deterministic variational approximation algorithm to learn the proposed probabilistic model. We then extend the proposed method to find intrinsic dimensionality of the projected subspace using automatic relevance determination and to handle semi-supervised learning using a low-density assumption. We perform supervised learning experiments on four benchmark multilabel learning data sets by comparing our method with baseline linear dimensionality reduction algorithms. These experiments show that the proposed approach achieves good performance values in terms of hamming loss, average AUC, macro F1, and micro F1 on held-out test data. The low-dimensional embeddings obtained by our method are also very useful for exploratory data analysis. We also show the effectiveness of our approach in finding intrinsic subspace dimensionality and semi-supervised learning tasks. PMID:24532862

Coupled dimensionality reduction and classification for supervised and semi-supervised multilabel learning.

PubMed

Gönen, Mehmet

2014-03-01

Coupled training of dimensionality reduction and classification is proposed previously to improve the prediction performance for single-label problems. Following this line of research, in this paper, we first introduce a novel Bayesian method that combines linear dimensionality reduction with linear binary classification for supervised multilabel learning and present a deterministic variational approximation algorithm to learn the proposed probabilistic model. We then extend the proposed method to find intrinsic dimensionality of the projected subspace using automatic relevance determination and to handle semi-supervised learning using a low-density assumption. We perform supervised learning experiments on four benchmark multilabel learning data sets by comparing our method with baseline linear dimensionality reduction algorithms. These experiments show that the proposed approach achieves good performance values in terms of hamming loss, average AUC, macro F 1 , and micro F 1 on held-out test data. The low-dimensional embeddings obtained by our method are also very useful for exploratory data analysis. We also show the effectiveness of our approach in finding intrinsic subspace dimensionality and semi-supervised learning tasks.

Accurate crop classification using hierarchical genetic fuzzy rule-based systems

NASA Astrophysics Data System (ADS)

Topaloglou, Charalampos A.; Mylonas, Stelios K.; Stavrakoudis, Dimitris G.; Mastorocostas, Paris A.; Theocharis, John B.

2014-10-01

This paper investigates the effectiveness of an advanced classification system for accurate crop classification using very high resolution (VHR) satellite imagery. Specifically, a recently proposed genetic fuzzy rule-based classification system (GFRBCS) is employed, namely, the Hierarchical Rule-based Linguistic Classifier (HiRLiC). HiRLiC's model comprises a small set of simple IF-THEN fuzzy rules, easily interpretable by humans. One of its most important attributes is that its learning algorithm requires minimum user interaction, since the most important learning parameters affecting the classification accuracy are determined by the learning algorithm automatically. HiRLiC is applied in a challenging crop classification task, using a SPOT5 satellite image over an intensively cultivated area in a lake-wetland ecosystem in northern Greece. A rich set of higher-order spectral and textural features is derived from the initial bands of the (pan-sharpened) image, resulting in an input space comprising 119 features. The experimental analysis proves that HiRLiC compares favorably to other interpretable classifiers of the literature, both in terms of structural complexity and classification accuracy. Its testing accuracy was very close to that obtained by complex state-of-the-art classification systems, such as the support vector machines (SVM) and random forest (RF) classifiers. Nevertheless, visual inspection of the derived classification maps shows that HiRLiC is characterized by higher generalization properties, providing more homogeneous classifications that the competitors. Moreover, the runtime requirements for producing the thematic map was orders of magnitude lower than the respective for the competitors.

Estimating persistence of brominated and chlorinated organic pollutants in air, water, soil, and sediments with the QSPR-based classification scheme.

PubMed

Puzyn, T; Haranczyk, M; Suzuki, N; Sakurai, T

2011-02-01

We have estimated degradation half-lives of both brominated and chlorinated dibenzo-p-dioxins (PBDDs and PCDDs), furans (PBDFs and PCDFs), biphenyls (PBBs and PCBs), naphthalenes (PBNs and PCNs), diphenyl ethers (PBDEs and PCDEs) as well as selected unsubstituted polycyclic aromatic hydrocarbons (PAHs) in air, surface water, surface soil, and sediments (in total of 1,431 compounds in four compartments). Next, we compared the persistence between chloro- (relatively well-studied) and bromo- (less studied) analogs. The predictions have been performed based on the quantitative structure-property relationship (QSPR) scheme with use of k-nearest neighbors (kNN) classifier and the semi-quantitative system of persistence classes. The classification models utilized principal components derived from the principal component analysis of a set of 24 constitutional and quantum mechanical descriptors as input variables. Accuracies of classification (based on an external validation) were 86, 85, 87, and 75% for air, surface water, surface soil, and sediments, respectively. The persistence of all chlorinated species increased with increasing halogenation degree. In the case of brominated organic pollutants (Br-OPs), the trend was the same for air and sediments. However, we noticed that the opposite trend for persistence in surface water and soil. The results suggest that, due to high photoreactivity of C-Br chemical bonds, photolytic processes occurring in surface water and soil are able to play significant role in transforming and removing Br-OPs from these compartments. This contribution is the first attempt of classifying together Br-OPs and Cl-OPs according to their persistence, in particular, environmental compartments.

Is the textural classification built on sand?

USDA-ARS?s Scientific Manuscript database

In 1967, the Committee of the Soil Science Society of America noted that the current system of particle size boundaries arose due to geographic accident. The committee noted that there is “no narrowly defineable natural particle size boundaries that would be equally significant in all soil materials...

A Geostatistical Toolset for Reconstructing Louisiana's Coastal Stratigraphy using Subsurface Boring and Cone Penetrometer Test Data

NASA Astrophysics Data System (ADS)

Li, A.; Tsai, F. T. C.; Jafari, N.; Chen, Q. J.; Bentley, S. J.

2017-12-01

A vast area of river deltaic wetlands stretches across southern Louisiana coast. The wetlands are suffering from a high rate of land loss, which increasingly threats coastal community and energy infrastructure. A regional stratigraphic framework of the delta plain is now imperative to answer scientific questions (such as how the delta plain grows and decays?) and to provide information to coastal protection and restoration projects (such as marsh creation and construction of levees and floodwalls). Through years, subsurface investigations in Louisiana have been conducted by state and federal agencies (Louisiana Department of Natural Resources, United States Geological Survey, United States Army Corps of Engineers, etc.), research institutes (Louisiana Geological Survey, LSU Coastal Studies Institute, etc.), engineering firms, and oil-gas companies. This has resulted in the availability of various types of data, including geological, geotechnical, and geophysical data. However, it is challenging to integrate different types of data and construct three-dimensional stratigraphy models in regional scale. In this study, a set of geostatistical methods were used to tackle this problem. An ordinary kriging method was used to regionalize continuous data, such as grain size, water content, liquid limit, plasticity index, and cone penetrometer tests (CPTs). Indicator kriging and multiple indicator kriging methods were used to regionalize categorized data, such as soil classification. A compositional kriging method was used to regionalize compositional data, such as soil composition (fractions of sand, silt and clay). Stratigraphy models were constructed for three cases in the coastal zone: (1) Inner Harbor Navigation Canal (IHNC) area: soil classification and soil behavior type (SBT) stratigraphies were constructed using ordinary kriging; (2) Middle Barataria Bay area: a soil classification stratigraphy was constructed using multiple indicator kriging; (3) Lower Barataria Bay and Lower Breton Sound areas: a soil texture stratigraphy was constructed using soil compositional data and compositional kriging. Cross sections were extracted from the three-dimensional stratigraphy models to reveal spatial distributions of different stratigraphic features.

A model for predicting embankment slope failures in clay-rich soils; A Louisiana example

NASA Astrophysics Data System (ADS)

Burns, S. F.

2015-12-01

A model for predicting embankment slope failures in clay-rich soils; A Louisiana example It is well known that smectite-rich soils significantly reduce the stability of slopes. The question is how much smectite in the soil causes slope failures. A study of over 100 sites in north and south Louisiana, USA, compared slopes that failed during a major El Nino winter (heavy rainfall) in 1982-1983 to similar slopes that did not fail. Soils in the slopes were tested for per cent clay, liquid limits, plasticity indices and semi-quantitative clay mineralogy. Slopes with the High Risk for failure (85-90% chance of failure in 8-15 years after construction) contained soils with a liquid limit > 54%, a plasticity index over 29%, and clay contents > 47%. Slopes with an Intermediate Risk (55-50% chance of failure in 8-15 years) contained soils with a liquid limit between 36-54%, plasticity index between 16-19%, and clay content between 32-47%. Slopes with a Low Risk chance of failure (< 5% chance of failure in 8-15 years after construction) contained soils with a liquid limit < 36%, a plasticity index < 16%, and a clay content < 32%. These data show that if one is constructing embankments and one wants to prevent slope failure of the 3:1 slopes, check the above soil characteristics before construction. If the soils fall into the Low Risk classification, construct the embankment normally. If the soils fall into the High Risk classification, one will need to use lime stabilization or heat treatments to prevent failures. Soils in the Intermediate Risk class will have to be evaluated on a case by case basis.

Sequence of Changes in Maize Responding to Soil Water Deficit and Related Critical Thresholds

PubMed Central

Ma, Xueyan; He, Qijin; Zhou, Guangsheng

2018-01-01

The sequence of changes in crop responding to soil water deficit and related critical thresholds are essential for better drought damage classification and drought monitoring indicators. This study was aimed to investigate the critical thresholds of maize growth and physiological characteristics responding to changing soil water and to reveal the sequence of changes in maize responding to soil water deficit both in seedling and jointing stages based on 2-year’s maize field experiment responding to six initial soil water statuses conducted in 2013 and 2014. Normal distribution tolerance limits were newly adopted to identify critical thresholds of maize growth and physiological characteristics to a wide range of soil water status. The results showed that in both stages maize growth characteristics related to plant water status [stem moisture content (SMC) and leaf moisture content (LMC)], leaf gas exchange [net photosynthetic rate (Pn), transpiration rate (Tr), and stomatal conductance (Gs)], and leaf area were sensitive to soil water deficit, while biomass-related characteristics were less sensitive. Under the concurrent weather conditions and agronomic managements, the critical soil water thresholds in terms of relative soil moisture of 0–30 cm depth (RSM) of maize SMC, LMC, net Pn, Tr, Gs, and leaf area were 72, 65, 62, 60, 58, and 46%, respectively, in seedling stage, and 64, 64, 51, 53, 48, and 46%, respectively, in jointing stage. It indicated that there is a sequence of changes in maize responding to soil water deficit, i.e., their response sequences as soil water deficit intensified: SMC ≥ LMC > leaf gas exchange > leaf area in both stages. This sequence of changes in maize responding to soil water deficit and related critical thresholds may be better indicators of damage classification and drought monitoring. PMID:29765381

Space Object Classification Using Fused Features of Time Series Data

NASA Astrophysics Data System (ADS)

Jia, B.; Pham, K. D.; Blasch, E.; Shen, D.; Wang, Z.; Chen, G.

In this paper, a fused feature vector consisting of raw time series and texture feature information is proposed for space object classification. The time series data includes historical orbit trajectories and asteroid light curves. The texture feature is derived from recurrence plots using Gabor filters for both unsupervised learning and supervised learning algorithms. The simulation results show that the classification algorithms using the fused feature vector achieve better performance than those using raw time series or texture features only.

A Robust Geometric Model for Argument Classification

NASA Astrophysics Data System (ADS)

Giannone, Cristina; Croce, Danilo; Basili, Roberto; de Cao, Diego

Argument classification is the task of assigning semantic roles to syntactic structures in natural language sentences. Supervised learning techniques for frame semantics have been recently shown to benefit from rich sets of syntactic features. However argument classification is also highly dependent on the semantics of the involved lexicals. Empirical studies have shown that domain dependence of lexical information causes large performance drops in outside domain tests. In this paper a distributional approach is proposed to improve the robustness of the learning model against out-of-domain lexical phenomena.

Dialogic and integrated approach to promote soils at different school levels: a Brazilian experience

NASA Astrophysics Data System (ADS)

Muggler, Cristine Carole

2017-04-01

From ancient civilizations to present technological societies, soil is the material and immaterial ground of our existence. Soil is essential to life as are water, air and sun light. Nevertheless, it is overlooked and has its functions and importance not known and recognized by people. In formal education and in most school curricula, soil contents are not approached in the same way and intensity other environmental components are. In its essence, soils are an interdisciplinary subject, crossing over different disciplines. It has a great potential as unifying theme that links and synthesizes different contents and areas of knowledge, especially hard sciences as physics, chemistry and biology. Furthermore, soils are familiar and tangible to everyone, making them a meaningful subject that helps to build an efficient learning process. The challenge remains on how to bring such teaching-learning possibilities to formal education at all levels. Soil education deals with the significance of soil to people. What makes soil meaningful? What are the bases for effective learning about soil? The answers are very much related with subjective perceptions and life experiences carried by each individual. Those dimensions have been considered by the pedagogical approach based on Paulo Freire's socio constructivism which considers social inclusion, knowledge building, horizontal learning and collective action. This approach has been applied within the soil (science) education spaces of the Federal University of Viçosa, Minas Gerais, Brazil, both with university students and basic education pupils. At the university an average of 200 students per semester follow a 60 hours Soil Genesis course. With primary and secondary schools the activities are developed through the Soil Education Programme (PES) of the Earth Sciences Museum. In the classes and activities, materials, methods and learning strategies are developed to stimulate involvement, dialogues and exchange of experiences and knowledge between students themselves and between students and teachers in order to build and re-build their understanding of soils. Those strategies include hands-on activities, field visits, landscape observations, collective productions and artistic works among other strategies. They are done in a dialogic and horizontal way where each ones' perceptions and experiences is valued and considered for the building of knowledge on soils. Good achievements have been obtained when university students are involved in outreach activities aimed to basic education schools and to general public, in a "teach to learn" approach.

Drug related webpages classification using images and text information based on multi-kernel learning

NASA Astrophysics Data System (ADS)

Hu, Ruiguang; Xiao, Liping; Zheng, Wenjuan

2015-12-01

In this paper, multi-kernel learning(MKL) is used for drug-related webpages classification. First, body text and image-label text are extracted through HTML parsing, and valid images are chosen by the FOCARSS algorithm. Second, text based BOW model is used to generate text representation, and image-based BOW model is used to generate images representation. Last, text and images representation are fused with a few methods. Experimental results demonstrate that the classification accuracy of MKL is higher than those of all other fusion methods in decision level and feature level, and much higher than the accuracy of single-modal classification.

Online clustering algorithms for radar emitter classification.

PubMed

Liu, Jun; Lee, Jim P Y; Senior; Li, Lingjie; Luo, Zhi-Quan; Wong, K Max

2005-08-01

Radar emitter classification is a special application of data clustering for classifying unknown radar emitters from received radar pulse samples. The main challenges of this task are the high dimensionality of radar pulse samples, small sample group size, and closely located radar pulse clusters. In this paper, two new online clustering algorithms are developed for radar emitter classification: One is model-based using the Minimum Description Length (MDL) criterion and the other is based on competitive learning. Computational complexity is analyzed for each algorithm and then compared. Simulation results show the superior performance of the model-based algorithm over competitive learning in terms of better classification accuracy, flexibility, and stability.

An efficient ensemble learning method for gene microarray classification.

PubMed

Osareh, Alireza; Shadgar, Bita

2013-01-01

The gene microarray analysis and classification have demonstrated an effective way for the effective diagnosis of diseases and cancers. However, it has been also revealed that the basic classification techniques have intrinsic drawbacks in achieving accurate gene classification and cancer diagnosis. On the other hand, classifier ensembles have received increasing attention in various applications. Here, we address the gene classification issue using RotBoost ensemble methodology. This method is a combination of Rotation Forest and AdaBoost techniques which in turn preserve both desirable features of an ensemble architecture, that is, accuracy and diversity. To select a concise subset of informative genes, 5 different feature selection algorithms are considered. To assess the efficiency of the RotBoost, other nonensemble/ensemble techniques including Decision Trees, Support Vector Machines, Rotation Forest, AdaBoost, and Bagging are also deployed. Experimental results have revealed that the combination of the fast correlation-based feature selection method with ICA-based RotBoost ensemble is highly effective for gene classification. In fact, the proposed method can create ensemble classifiers which outperform not only the classifiers produced by the conventional machine learning but also the classifiers generated by two widely used conventional ensemble learning methods, that is, Bagging and AdaBoost.

Cell dynamic morphology classification using deep convolutional neural networks.

PubMed

Li, Heng; Pang, Fengqian; Shi, Yonggang; Liu, Zhiwen

2018-05-15

Cell morphology is often used as a proxy measurement of cell status to understand cell physiology. Hence, interpretation of cell dynamic morphology is a meaningful task in biomedical research. Inspired by the recent success of deep learning, we here explore the application of convolutional neural networks (CNNs) to cell dynamic morphology classification. An innovative strategy for the implementation of CNNs is introduced in this study. Mouse lymphocytes were collected to observe the dynamic morphology, and two datasets were thus set up to investigate the performances of CNNs. Considering the installation of deep learning, the classification problem was simplified from video data to image data, and was then solved by CNNs in a self-taught manner with the generated image data. CNNs were separately performed in three installation scenarios and compared with existing methods. Experimental results demonstrated the potential of CNNs in cell dynamic morphology classification, and validated the effectiveness of the proposed strategy. CNNs were successfully applied to the classification problem, and outperformed the existing methods in the classification accuracy. For the installation of CNNs, transfer learning was proved to be a promising scheme. © 2018 International Society for Advancement of Cytometry. © 2018 International Society for Advancement of Cytometry.

General functioning predicts reward and punishment learning in schizophrenia.

PubMed

Somlai, Zsuzsanna; Moustafa, Ahmed A; Kéri, Szabolcs; Myers, Catherine E; Gluck, Mark A

2011-04-01

Previous studies investigating feedback-driven reinforcement learning in patients with schizophrenia have provided mixed results. In this study, we explored the clinical predictors of reward and punishment learning using a probabilistic classification learning task. Patients with schizophrenia (n=40) performed similarly to healthy controls (n=30) on the classification learning task. However, more severe negative and general symptoms were associated with lower reward-learning performance, whereas poorer general psychosocial functioning was correlated with both lower reward- and punishment-learning performances. Multiple linear regression analyses indicated that general psychosocial functioning was the only significant predictor of reinforcement learning performance when education, antipsychotic dose, and positive, negative and general symptoms were included in the analysis. These results suggest a close relationship between reinforcement learning and general psychosocial functioning in schizophrenia. Published by Elsevier B.V.

Classifying EEG for Brain-Computer Interface: Learning Optimal Filters for Dynamical System Features

PubMed Central

Song, Le; Epps, Julien

2007-01-01

Classification of multichannel EEG recordings during motor imagination has been exploited successfully for brain-computer interfaces (BCI). In this paper, we consider EEG signals as the outputs of a networked dynamical system (the cortex), and exploit synchronization features from the dynamical system for classification. Herein, we also propose a new framework for learning optimal filters automatically from the data, by employing a Fisher ratio criterion. Experimental evaluations comparing the proposed dynamical system features with the CSP and the AR features reveal their competitive performance during classification. Results also show the benefits of employing the spatial and the temporal filters optimized using the proposed learning approach. PMID:18364986

Diverse Region-Based CNN for Hyperspectral Image Classification.

PubMed

Zhang, Mengmeng; Li, Wei; Du, Qian

2018-06-01

Convolutional neural network (CNN) is of great interest in machine learning and has demonstrated excellent performance in hyperspectral image classification. In this paper, we propose a classification framework, called diverse region-based CNN, which can encode semantic context-aware representation to obtain promising features. With merging a diverse set of discriminative appearance factors, the resulting CNN-based representation exhibits spatial-spectral context sensitivity that is essential for accurate pixel classification. The proposed method exploiting diverse region-based inputs to learn contextual interactional features is expected to have more discriminative power. The joint representation containing rich spectral and spatial information is then fed to a fully connected network and the label of each pixel vector is predicted by a softmax layer. Experimental results with widely used hyperspectral image data sets demonstrate that the proposed method can surpass any other conventional deep learning-based classifiers and other state-of-the-art classifiers.

Cell classification using big data analytics plus time stretch imaging (Conference Presentation)

NASA Astrophysics Data System (ADS)

Jalali, Bahram; Chen, Claire L.; Mahjoubfar, Ata

2016-09-01

We show that blood cells can be classified with high accuracy and high throughput by combining machine learning with time stretch quantitative phase imaging. Our diagnostic system captures quantitative phase images in a flow microscope at millions of frames per second and extracts multiple biophysical features from individual cells including morphological characteristics, light absorption and scattering parameters, and protein concentration. These parameters form a hyperdimensional feature space in which supervised learning and cell classification is performed. We show binary classification of T-cells against colon cancer cells, as well classification of algae cell strains with high and low lipid content. The label-free screening averts the negative impact of staining reagents on cellular viability or cell signaling. The combination of time stretch machine vision and learning offers unprecedented cell analysis capabilities for cancer diagnostics, drug development and liquid biopsy for personalized genomics.

Alexnet Feature Extraction and Multi-Kernel Learning for Objectoriented Classification

NASA Astrophysics Data System (ADS)

Ding, L.; Li, H.; Hu, C.; Zhang, W.; Wang, S.

2018-04-01

In view of the fact that the deep convolutional neural network has stronger ability of feature learning and feature expression, an exploratory research is done on feature extraction and classification for high resolution remote sensing images. Taking the Google image with 0.3 meter spatial resolution in Ludian area of Yunnan Province as an example, the image segmentation object was taken as the basic unit, and the pre-trained AlexNet deep convolution neural network model was used for feature extraction. And the spectral features, AlexNet features and GLCM texture features are combined with multi-kernel learning and SVM classifier, finally the classification results were compared and analyzed. The results show that the deep convolution neural network can extract more accurate remote sensing image features, and significantly improve the overall accuracy of classification, and provide a reference value for earthquake disaster investigation and remote sensing disaster evaluation.

Classification of EEG signals using a genetic-based machine learning classifier.

PubMed

Skinner, B T; Nguyen, H T; Liu, D K

2007-01-01

This paper investigates the efficacy of the genetic-based learning classifier system XCS, for the classification of noisy, artefact-inclusive human electroencephalogram (EEG) signals represented using large condition strings (108bits). EEG signals from three participants were recorded while they performed four mental tasks designed to elicit hemispheric responses. Autoregressive (AR) models and Fast Fourier Transform (FFT) methods were used to form feature vectors with which mental tasks can be discriminated. XCS achieved a maximum classification accuracy of 99.3% and a best average of 88.9%. The relative classification performance of XCS was then compared against four non-evolutionary classifier systems originating from different learning techniques. The experimental results will be used as part of our larger research effort investigating the feasibility of using EEG signals as an interface to allow paralysed persons to control a powered wheelchair or other devices.

Classification of highly unbalanced CYP450 data of drugs using cost sensitive machine learning techniques.

PubMed

Eitrich, T; Kless, A; Druska, C; Meyer, W; Grotendorst, J

2007-01-01

In this paper, we study the classifications of unbalanced data sets of drugs. As an example we chose a data set of 2D6 inhibitors of cytochrome P450. The human cytochrome P450 2D6 isoform plays a key role in the metabolism of many drugs in the preclinical drug discovery process. We have collected a data set from annotated public data and calculated physicochemical properties with chemoinformatics methods. On top of this data, we have built classifiers based on machine learning methods. Data sets with different class distributions lead to the effect that conventional machine learning methods are biased toward the larger class. To overcome this problem and to obtain sensitive but also accurate classifiers we combine machine learning and feature selection methods with techniques addressing the problem of unbalanced classification, such as oversampling and threshold moving. We have used our own implementation of a support vector machine algorithm as well as the maximum entropy method. Our feature selection is based on the unsupervised McCabe method. The classification results from our test set are compared structurally with compounds from the training set. We show that the applied algorithms enable the effective high throughput in silico classification of potential drug candidates.

Detection of Periodic Leg Movements by Machine Learning Methods Using Polysomnographic Parameters Other Than Leg Electromyography

PubMed Central

Umut, İlhan; Çentik, Güven

2016-01-01

The number of channels used for polysomnographic recording frequently causes difficulties for patients because of the many cables connected. Also, it increases the risk of having troubles during recording process and increases the storage volume. In this study, it is intended to detect periodic leg movement (PLM) in sleep with the use of the channels except leg electromyography (EMG) by analysing polysomnography (PSG) data with digital signal processing (DSP) and machine learning methods. PSG records of 153 patients of different ages and genders with PLM disorder diagnosis were examined retrospectively. A novel software was developed for the analysis of PSG records. The software utilizes the machine learning algorithms, statistical methods, and DSP methods. In order to classify PLM, popular machine learning methods (multilayer perceptron, K-nearest neighbour, and random forests) and logistic regression were used. Comparison of classified results showed that while K-nearest neighbour classification algorithm had higher average classification rate (91.87%) and lower average classification error value (RMSE = 0.2850), multilayer perceptron algorithm had the lowest average classification rate (83.29%) and the highest average classification error value (RMSE = 0.3705). Results showed that PLM can be classified with high accuracy (91.87%) without leg EMG record being present. PMID:27213008

Detection of Periodic Leg Movements by Machine Learning Methods Using Polysomnographic Parameters Other Than Leg Electromyography.

PubMed

Umut, İlhan; Çentik, Güven

2016-01-01

The number of channels used for polysomnographic recording frequently causes difficulties for patients because of the many cables connected. Also, it increases the risk of having troubles during recording process and increases the storage volume. In this study, it is intended to detect periodic leg movement (PLM) in sleep with the use of the channels except leg electromyography (EMG) by analysing polysomnography (PSG) data with digital signal processing (DSP) and machine learning methods. PSG records of 153 patients of different ages and genders with PLM disorder diagnosis were examined retrospectively. A novel software was developed for the analysis of PSG records. The software utilizes the machine learning algorithms, statistical methods, and DSP methods. In order to classify PLM, popular machine learning methods (multilayer perceptron, K-nearest neighbour, and random forests) and logistic regression were used. Comparison of classified results showed that while K-nearest neighbour classification algorithm had higher average classification rate (91.87%) and lower average classification error value (RMSE = 0.2850), multilayer perceptron algorithm had the lowest average classification rate (83.29%) and the highest average classification error value (RMSE = 0.3705). Results showed that PLM can be classified with high accuracy (91.87%) without leg EMG record being present.

Nondestructive Evaluation of Airport Pavements. Volume I. Program References,

DTIC Science & Technology

1979-09-01

greater than its original capacity (see test 13 on Fig. 2.5). During the material tests by Majidzadeh, the dynamic E-value of frozen subgrade soil was...Sample the base and subbase material by conventional spoon and identify the material by standard soil -aggregate classification and penetration...such as shaker table. The new testing specification is designed for all paving materials including subgrade soils . The specifications of material

Soil erosion modelling for NSW coastal catchments using RUSLE in a GIS environment

NASA Astrophysics Data System (ADS)

Yang, Xihua; Chapman, Greg

2006-10-01

In this study, hillslope erosion risk has been estimated for all eastern New South Wales (NSW) catchments, Australia using Revised Universal Soil Loss Equation (RUSLE) in a geographic information system (GIS) environment. Rainfall-runoff erosivity (R) factor was interpolated from NSW rainfall-erosivity contour (isoerodent) data. Soil erodibility (K) factor was based on the soil regolith stability and sediment yield classification. The classification was derived from soil landscape and related soil map data. The slope length and steepness (LS) factor was derived from high resolution digital elevation model (DEM). A fully-automated program using Arc Macro Language (AML) produced RUSLE-based LS factor grids for all coastal catchments. The outputs are comparable to the range of LS values summarised in the literature. Cover and management (C) factor and conservation support-practices (P) factor were set to one. They are intended to be allocated according to land use, ground cover and erosion control provisions for particular land management actions. The resulting erosion risk map, with pixel size of 25-m, provides unprecedented resolution of relative expected sheet and rill erosion across all NSW costal catchments and can be adapted for a range of erosion control purposes such as bushfire hazard reduction and comprehensive costal assessment.

Classification and Identification of Reading and Math Disabilities: The Special Case of Comorbidity

ERIC Educational Resources Information Center

Branum-Martin, Lee; Fletcher, Jack M.; Stuebing, Karla K.

2013-01-01

Much of learning disabilities research relies on categorical classification frameworks that use psychometric tests and cut points to identify children with reading or math difficulties. However, there is increasing evidence that the attributes of reading and math learning disabilities are dimensional, representing correlated continua of severity.…

Classification and evaluation for forest sites on the Southern Cumberland Plateau

Treesearch

Glendon W. Smalley

1979-01-01

This paper presents a comprehensive forest site classification system for the southern portion of the Cumberland Plateau in northern Alabama, northwest Georgia, and extreme south-central Tennessee. The system is based on physiography, geology, soils, topography, and vegetation. Twenty-one landtypes are described, and each landtype is evaluated in terms of productivity...

Fire severity classification: Uses and abuses

Treesearch

Theresa B. Jain; Russell T. Graham

2003-01-01

Burn severity (also referred to as fire severity) is not a single definition, but rather a concept and its classification is a function of the measured units unique to the system of interest. The systems include: flora and fauna, soil microbiology and hydrologic processes, atmospheric inputs, fire management, and society. Depending on the particular system of interest...

Predicting fire severity using surface fuels and moisture

Treesearch

Pamela G. Sikkink; Robert E. Keane

2012-01-01

Fire severity classifications have been used extensively in fire management over the last 30 years to describe specific environmental or ecological impacts of fire on fuels, vegetation, wildlife, and soils in recently burned areas. New fire severity classifications need to be more objective, predictive, and ultimately more useful to fire management and planning. Our...

Plant community classification for alpine vegetation on the Beaverhead National Forest, Montana

Treesearch

Stephen V. Cooper; Peter Lesica; Deborah Page-Dumroese

1997-01-01

Vegetation of the alpine zone of eight mountain ranges in southwestern Montana was classified using IWINSPAN, DECORAN, and STRATA-algorithms embedded within the U.S. Forest Service Northern Region's ECADS (ecological classification and description system) program. Quantitative estimates of vegetation and soil attributes were sampled from 138 plots. Vegetation...

Clay stabilization by using gypsum and paddy husk ash with reference to UCT and CBR value

NASA Astrophysics Data System (ADS)

Roesyanto; Iskandar, R.; Hastuty, I. P.; Dianty, W. O.

2018-02-01

Clays that have low shear strength need to be stabilized in order to meet the technical requirements to serve as a subgrade material. One of the usual soil stabilization methods is by adding chemicals such as Portland cement, lime, and bitumen. The clay stabilization research was done by adding gypsum and paddy husk ash. The research goals were to find out the value of engineering properties of clay due to the addition of 2% gypsum and 2% - 15% paddy husk ash. The soil was classified as Clay - Low Plasticity (CL) based on USCS and was classified as A-7-6 (10) based on AASHTO classification system. The UCT value of original soil was 1.41 kg/cm2. While the CBR soaked and unsoaked values of original soil were 4.41% and 6.23% respectively. The research results showed the addition of paddy husk ash decreased the value of unconfined compressive strength as well as CBR. The stabilized soil by 2% gypsum and 0% paddy husk ash gave maximum UCT value of 1.67 kg/cm2, while the maximum value of CBR were found 6.71% for CBR soaked and 8.00% for CBR unsoaked. The addition of paddy husk ash did not alter the soil classification according to AASHTO or USCS, even degrade the engineering properties of original soil.

Expected energy-based restricted Boltzmann machine for classification.

PubMed

Elfwing, S; Uchibe, E; Doya, K

2015-04-01

In classification tasks, restricted Boltzmann machines (RBMs) have predominantly been used in the first stage, either as feature extractors or to provide initialization of neural networks. In this study, we propose a discriminative learning approach to provide a self-contained RBM method for classification, inspired by free-energy based function approximation (FE-RBM), originally proposed for reinforcement learning. For classification, the FE-RBM method computes the output for an input vector and a class vector by the negative free energy of an RBM. Learning is achieved by stochastic gradient-descent using a mean-squared error training objective. In an earlier study, we demonstrated that the performance and the robustness of FE-RBM function approximation can be improved by scaling the free energy by a constant that is related to the size of network. In this study, we propose that the learning performance of RBM function approximation can be further improved by computing the output by the negative expected energy (EE-RBM), instead of the negative free energy. To create a deep learning architecture, we stack several RBMs on top of each other. We also connect the class nodes to all hidden layers to try to improve the performance even further. We validate the classification performance of EE-RBM using the MNIST data set and the NORB data set, achieving competitive performance compared with other classifiers such as standard neural networks, deep belief networks, classification RBMs, and support vector machines. The purpose of using the NORB data set is to demonstrate that EE-RBM with binary input nodes can achieve high performance in the continuous input domain. Copyright © 2014 The Authors. Published by Elsevier Ltd.. All rights reserved.

Comprehensive decision tree models in bioinformatics.

PubMed

Stiglic, Gregor; Kocbek, Simon; Pernek, Igor; Kokol, Peter

2012-01-01

Classification is an important and widely used machine learning technique in bioinformatics. Researchers and other end-users of machine learning software often prefer to work with comprehensible models where knowledge extraction and explanation of reasoning behind the classification model are possible. This paper presents an extension to an existing machine learning environment and a study on visual tuning of decision tree classifiers. The motivation for this research comes from the need to build effective and easily interpretable decision tree models by so called one-button data mining approach where no parameter tuning is needed. To avoid bias in classification, no classification performance measure is used during the tuning of the model that is constrained exclusively by the dimensions of the produced decision tree. The proposed visual tuning of decision trees was evaluated on 40 datasets containing classical machine learning problems and 31 datasets from the field of bioinformatics. Although we did not expected significant differences in classification performance, the results demonstrate a significant increase of accuracy in less complex visually tuned decision trees. In contrast to classical machine learning benchmarking datasets, we observe higher accuracy gains in bioinformatics datasets. Additionally, a user study was carried out to confirm the assumption that the tree tuning times are significantly lower for the proposed method in comparison to manual tuning of the decision tree. The empirical results demonstrate that by building simple models constrained by predefined visual boundaries, one not only achieves good comprehensibility, but also very good classification performance that does not differ from usually more complex models built using default settings of the classical decision tree algorithm. In addition, our study demonstrates the suitability of visually tuned decision trees for datasets with binary class attributes and a high number of possibly redundant attributes that are very common in bioinformatics.

Comprehensive Decision Tree Models in Bioinformatics

PubMed Central

Stiglic, Gregor; Kocbek, Simon; Pernek, Igor; Kokol, Peter

2012-01-01

Purpose Classification is an important and widely used machine learning technique in bioinformatics. Researchers and other end-users of machine learning software often prefer to work with comprehensible models where knowledge extraction and explanation of reasoning behind the classification model are possible. Methods This paper presents an extension to an existing machine learning environment and a study on visual tuning of decision tree classifiers. The motivation for this research comes from the need to build effective and easily interpretable decision tree models by so called one-button data mining approach where no parameter tuning is needed. To avoid bias in classification, no classification performance measure is used during the tuning of the model that is constrained exclusively by the dimensions of the produced decision tree. Results The proposed visual tuning of decision trees was evaluated on 40 datasets containing classical machine learning problems and 31 datasets from the field of bioinformatics. Although we did not expected significant differences in classification performance, the results demonstrate a significant increase of accuracy in less complex visually tuned decision trees. In contrast to classical machine learning benchmarking datasets, we observe higher accuracy gains in bioinformatics datasets. Additionally, a user study was carried out to confirm the assumption that the tree tuning times are significantly lower for the proposed method in comparison to manual tuning of the decision tree. Conclusions The empirical results demonstrate that by building simple models constrained by predefined visual boundaries, one not only achieves good comprehensibility, but also very good classification performance that does not differ from usually more complex models built using default settings of the classical decision tree algorithm. In addition, our study demonstrates the suitability of visually tuned decision trees for datasets with binary class attributes and a high number of possibly redundant attributes that are very common in bioinformatics. PMID:22479449

Learn More

Science.gov Websites

Learn More Related Books for Adults Cohen, Benjamin. Notes from the Ground: Science, Soil & Society in the American Countryside. Hillel, Daniel. Out of the Earth: Civilization and the Life of the Soil Civilizations. Van Beek, Gus and Ora. Glorious Mud! Related Books for Children Wermund, Jerry. Soil: More Than

Crossword Puzzles as Learning Tools in Introductory Soil Science

ERIC Educational Resources Information Center

Barbarick, K. A.

2010-01-01

Students in introductory courses generally respond favorably to novel approaches to learning. To this end, I developed and used three crossword puzzles in spring and fall 2009 semesters in Introductory Soil Science Laboratory at Colorado State University. The first hypothesis was that crossword puzzles would improve introductory soil science…

Influence of Fines Content on Consolidation and Compressibility Characteristics of Granular Materials

NASA Astrophysics Data System (ADS)

Lipiński, Mirosław J.; Wdowska, Małgorzata K.; Jaroń, Łukasz

2017-10-01

Various behaviour of soil under loading results to large extent from kind of soil considered. There is a lot of literature concerning pure sand or plastic clays, while little is known about materials, which are from classification point of view, between those soils. These materials can be considered as cohesionless soils with various fines content. The paper present results of tests carried out in large consolidometer on three kinds of soil, containing 10, 36 and 97% of fines content. Consolidation, permeability and compressibility characteristics were determined. Analysis of the test results allowed to formulate conclusion concerning change in soil behaviour resulting from fines content.

A machine learning approach for viral genome classification.

PubMed

Remita, Mohamed Amine; Halioui, Ahmed; Malick Diouara, Abou Abdallah; Daigle, Bruno; Kiani, Golrokh; Diallo, Abdoulaye Baniré

2017-04-11

Advances in cloning and sequencing technology are yielding a massive number of viral genomes. The classification and annotation of these genomes constitute important assets in the discovery of genomic variability, taxonomic characteristics and disease mechanisms. Existing classification methods are often designed for specific well-studied family of viruses. Thus, the viral comparative genomic studies could benefit from more generic, fast and accurate tools for classifying and typing newly sequenced strains of diverse virus families. Here, we introduce a virus classification platform, CASTOR, based on machine learning methods. CASTOR is inspired by a well-known technique in molecular biology: restriction fragment length polymorphism (RFLP). It simulates, in silico, the restriction digestion of genomic material by different enzymes into fragments. It uses two metrics to construct feature vectors for machine learning algorithms in the classification step. We benchmark CASTOR for the classification of distinct datasets of human papillomaviruses (HPV), hepatitis B viruses (HBV) and human immunodeficiency viruses type 1 (HIV-1). Results reveal true positive rates of 99%, 99% and 98% for HPV Alpha species, HBV genotyping and HIV-1 M subtyping, respectively. Furthermore, CASTOR shows a competitive performance compared to well-known HIV-1 specific classifiers (REGA and COMET) on whole genomes and pol fragments. The performance of CASTOR, its genericity and robustness could permit to perform novel and accurate large scale virus studies. The CASTOR web platform provides an open access, collaborative and reproducible machine learning classifiers. CASTOR can be accessed at http://castor.bioinfo.uqam.ca .

Long-range dismount activity classification: LODAC

NASA Astrophysics Data System (ADS)

Garagic, Denis; Peskoe, Jacob; Liu, Fang; Cuevas, Manuel; Freeman, Andrew M.; Rhodes, Bradley J.

2014-06-01

Continuous classification of dismount types (including gender, age, ethnicity) and their activities (such as walking, running) evolving over space and time is challenging. Limited sensor resolution (often exacerbated as a function of platform standoff distance) and clutter from shadows in dense target environments, unfavorable environmental conditions, and the normal properties of real data all contribute to the challenge. The unique and innovative aspect of our approach is a synthesis of multimodal signal processing with incremental non-parametric, hierarchical Bayesian machine learning methods to create a new kind of target classification architecture. This architecture is designed from the ground up to optimally exploit correlations among the multiple sensing modalities (multimodal data fusion) and rapidly and continuously learns (online self-tuning) patterns of distinct classes of dismounts given little a priori information. This increases classification performance in the presence of challenges posed by anti-access/area denial (A2/AD) sensing. To fuse multimodal features, Long-range Dismount Activity Classification (LODAC) develops a novel statistical information theoretic approach for multimodal data fusion that jointly models multimodal data (i.e., a probabilistic model for cross-modal signal generation) and discovers the critical cross-modal correlations by identifying components (features) with maximal mutual information (MI) which is efficiently estimated using non-parametric entropy models. LODAC develops a generic probabilistic pattern learning and classification framework based on a new class of hierarchical Bayesian learning algorithms for efficiently discovering recurring patterns (classes of dismounts) in multiple simultaneous time series (sensor modalities) at multiple levels of feature granularity.

GIS/RS-based Rapid Reassessment for Slope Land Capability Classification

NASA Astrophysics Data System (ADS)

Chang, T. Y.; Chompuchan, C.

2014-12-01

Farmland resources in Taiwan are limited because about 73% is mountainous and slope land. Moreover, the rapid urbanization and dense population resulted in the highly developed flat area. Therefore, the utilization of slope land for agriculture is more needed. In 1976, "Slope Land Conservation and Utilization Act" was promulgated to regulate the slope land utilization. Consequently, slope land capability was categorized into Class I-IV according to 4 criteria, i.e., average land slope, effective soil depth, degree of soil erosion, and parent rock. The slope land capability Class I-VI are suitable for cultivation and pasture. Whereas, Class V should be used for forestry purpose and Class VI should be the conservation land which requires intensive conservation practices. The field survey was conducted to categorize each land unit as the classification scheme. The landowners may not allow to overuse land capability limitation. In the last decade, typhoons and landslides frequently devastated in Taiwan. The rapid post-disaster reassessment of the slope land capability classification is necessary. However, the large-scale disaster on slope land is the constraint of field investigation. This study focused on using satellite remote sensing and GIS as the rapid re-evaluation method. Chenyulan watershed in Nantou County, Taiwan was selected to be a case study area. Grid-based slope derivation, topographic wetness index (TWI) and USLE soil loss calculation were used to classify slope land capability. The results showed that GIS-based classification give an overall accuracy of 68.32%. In addition, the post-disaster areas of Typhoon Morakot in 2009, which interpreted by SPOT satellite imageries, were suggested to classify as the conservation lands. These tools perform better in the large coverage post-disaster update for slope land capability classification and reduce time-consuming, manpower and material resources to the field investigation.

Tools based on multivariate statistical analysis for classification of soil and groundwater in Apulian agricultural sites.

PubMed

Ielpo, Pierina; Leardi, Riccardo; Pappagallo, Giuseppe; Uricchio, Vito Felice

2017-06-01

In this paper, the results obtained from multivariate statistical techniques such as PCA (Principal component analysis) and LDA (Linear discriminant analysis) applied to a wide soil data set are presented. The results have been compared with those obtained on a groundwater data set, whose samples were collected together with soil ones, within the project "Improvement of the Regional Agro-meteorological Monitoring Network (2004-2007)". LDA, applied to soil data, has allowed to distinguish the geographical origin of the sample from either one of the two macroaeras: Bari and Foggia provinces vs Brindisi, Lecce e Taranto provinces, with a percentage of correct prediction in cross validation of 87%. In the case of the groundwater data set, the best classification was obtained when the samples were grouped into three macroareas: Foggia province, Bari province and Brindisi, Lecce and Taranto provinces, by reaching a percentage of correct predictions in cross validation of 84%. The obtained information can be very useful in supporting soil and water resource management, such as the reduction of water consumption and the reduction of energy and chemical (nutrients and pesticides) inputs in agriculture.

A Study of Hand Back Skin Texture Patterns for Personal Identification and Gender Classification

PubMed Central

Xie, Jin; Zhang, Lei; You, Jane; Zhang, David; Qu, Xiaofeng

2012-01-01

Human hand back skin texture (HBST) is often consistent for a person and distinctive from person to person. In this paper, we study the HBST pattern recognition problem with applications to personal identification and gender classification. A specially designed system is developed to capture HBST images, and an HBST image database was established, which consists of 1,920 images from 80 persons (160 hands). An efficient texton learning based method is then presented to classify the HBST patterns. First, textons are learned in the space of filter bank responses from a set of training images using the l1 -minimization based sparse representation (SR) technique. Then, under the SR framework, we represent the feature vector at each pixel over the learned dictionary to construct a representation coefficient histogram. Finally, the coefficient histogram is used as skin texture feature for classification. Experiments on personal identification and gender classification are performed by using the established HBST database. The results show that HBST can be used to assist human identification and gender classification. PMID:23012512

Estimating of Soil Texture Using Landsat Imagery: a Case Study in Thatta Tehsil, Sindh

NASA Astrophysics Data System (ADS)

Khalil, Zahid

2016-07-01

Soil texture is considered as an important environment factor for agricultural growth. It is the most essential part for soil classification in large scale. Today the precise soil information in large scale is of great demand from various stakeholders including soil scientists, environmental managers, land use planners and traditional agricultural users. With the increasing demand of soil properties in fine scale spatial resolution made the traditional laboratory methods inadequate. In addition the costs of soil analysis with precision agriculture systems are more expensive than traditional methods. In this regard, the application of geo-spatial techniques can be used as an alternative for examining soil analysis. This study aims to examine the ability of Geo-spatial techniques in identifying the spatial patterns of soil attributes in fine scale. Around 28 samples of soil were collected from the different areas of Thatta Tehsil, Sindh, Pakistan for analyzing soil texture. An Ordinary Least Square (OLS) regression analysis was used to relate the reflectance values of Landsat8 OLI imagery with the soil variables. The analysis showed there was a significant relationship (p<0.05) of band 2 and 5 with silt% (R2 = 0.52), and band 4 and 6 with clay% (R2 =0.40). The equation derived from OLS analysis was then used for the whole study area for deriving soil attributes. The USDA textural classification triangle was implementing for the derivation of soil texture map in GIS environment. The outcome revealed that the 'sandy loam' was in great quantity followed by loam, sandy clay loam and clay loam. The outcome shows that the Geo-spatial techniques could be used efficiently for mapping soil texture of a larger area in fine scale. This technology helped in decreasing cost, time and increase detailed information by reducing field work to a considerable level.

The creation of digital thematic soil maps at the regional level (with the map of soil carbon pools in the Usa River basin as an example)

NASA Astrophysics Data System (ADS)

Pastukhov, A. V.; Kaverin, D. A.; Shchanov, V. M.

2016-09-01

A digital map of soil carbon pools was created for the forest-tundra ecotone in the Usa River basin with the use of ERDAS Imagine 2014 and ArcGIS 10.2 software. Supervised classification and thematic interpretation of satellite images and digital terrain models with the use of a georeferenced database on soil profiles were applied. Expert assessment of the natural diversity and representativeness of random samples for different soil groups was performed, and the minimal necessary size of the statistical sample was determined.

Information Forests

DTIC Science & Technology

2014-01-01

Classification confidence, or informative content of the subsets, is quantified by the Information Divergence. Our approach relates to active learning , semi-supervised learning, mixed generative/discriminative learning.

Identification and classification of structural soil conservation measures based on very high resolution stereo satellite data.

PubMed

Eckert, Sandra; Tesfay Ghebremicael, Selamawit; Hurni, Hans; Kohler, Thomas

2017-05-15

Land degradation affects large areas of land around the globe, with grave consequences for those living off the land. Major efforts are being made to implement soil and water conservation measures that counteract soil erosion and help secure vital ecosystem services. However, where and to what extent such measures have been implemented is often not well documented. Knowledge about this could help to identify areas where soil and water conservation measures are successfully supporting sustainable land management, as well as areas requiring urgent rehabilitation of conservation structures such as terraces and bunds. This study explores the potential of the latest satellite-based remote sensing technology for use in assessing and monitoring the extent of existing soil and water conservation structures. We used a set of very high resolution stereo Geoeye-1 satellite data, from which we derived a detailed digital surface model as well as a set of other spectral, terrain, texture, and filtered information layers. We developed and applied an object-based classification approach, working on two segmentation levels. On the coarser level, the aim was to delimit certain landscape zones. Information about these landscape zones is useful in distinguishing different types of soil and water conservation structures, as each zone contains certain specific types of structures. On the finer level, the goal was to extract and identify different types of linear soil and water conservation structures. The classification rules were based mainly on spectral, textural, shape, and topographic properties, and included object relationships. This approach enabled us to identify and separate from other classes the majority (78.5%) of terraces and bunds, as well as most hillside terraces (81.25%). Omission and commission errors are similar to those obtained by the few existing studies focusing on the same research objective but using different types of remotely sensed data. Based on our results, we estimate that the construction of the conservation structures in our study area in Eritrea required over 300,000 person-days of work, which underlines the huge efforts involved in soil and water conservation. Copyright © 2017 Elsevier Ltd. All rights reserved.

Geotechnical soil characterization of intact Quaternary deposits forming the March 22, 2014 SR-530 (Oso) landslide, Snohomish County, Washington

USGS Publications Warehouse

Riemer, Michael F.; Collins, Brian D.; Badger, Thomas C.; Toth, Csilla; Yu, Yat Chun

2015-01-01

This report provides a description of the methods used to obtain and test the intact soil stratigraphy behind the headscarp of the March 22 landslide. Detailed geotechnical index testing results are presented for 24 soil samples representing the stratigraphy at 19 different depths along a 650 ft (198 m) soil profile. The results include (1) the soil's in situ water content and unit weight (where applicable); (2) specific gravity of soil solids; and (3) each sample's grain-size distribution, critical limits for fine-grain water content states (that is, the Atterberg limits), and official Unified Soil Classification System (USCS) designation. In addition, preliminary stratigraphy and geotechnical relations within and between soil units are presented.

Photometric Supernova Classification with Machine Learning

NASA Astrophysics Data System (ADS)

Lochner, Michelle; McEwen, Jason D.; Peiris, Hiranya V.; Lahav, Ofer; Winter, Max K.

2016-08-01

Automated photometric supernova classification has become an active area of research in recent years in light of current and upcoming imaging surveys such as the Dark Energy Survey (DES) and the Large Synoptic Survey Telescope, given that spectroscopic confirmation of type for all supernovae discovered will be impossible. Here, we develop a multi-faceted classification pipeline, combining existing and new approaches. Our pipeline consists of two stages: extracting descriptive features from the light curves and classification using a machine learning algorithm. Our feature extraction methods vary from model-dependent techniques, namely SALT2 fits, to more independent techniques that fit parametric models to curves, to a completely model-independent wavelet approach. We cover a range of representative machine learning algorithms, including naive Bayes, k-nearest neighbors, support vector machines, artificial neural networks, and boosted decision trees (BDTs). We test the pipeline on simulated multi-band DES light curves from the Supernova Photometric Classification Challenge. Using the commonly used area under the curve (AUC) of the Receiver Operating Characteristic as a metric, we find that the SALT2 fits and the wavelet approach, with the BDTs algorithm, each achieve an AUC of 0.98, where 1 represents perfect classification. We find that a representative training set is essential for good classification, whatever the feature set or algorithm, with implications for spectroscopic follow-up. Importantly, we find that by using either the SALT2 or the wavelet feature sets with a BDT algorithm, accurate classification is possible purely from light curve data, without the need for any redshift information.

Classifying Acute Ischemic Stroke Onset Time using Deep Imaging Features

PubMed Central

Ho, King Chung; Speier, William; El-Saden, Suzie; Arnold, Corey W.

2017-01-01

Models have been developed to predict stroke outcomes (e.g., mortality) in attempt to provide better guidance for stroke treatment. However, there is little work in developing classification models for the problem of unknown time-since-stroke (TSS), which determines a patient’s treatment eligibility based on a clinical defined cutoff time point (i.e., <4.5hrs). In this paper, we construct and compare machine learning methods to classify TSS<4.5hrs using magnetic resonance (MR) imaging features. We also propose a deep learning model to extract hidden representations from the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional imaging features. Finally, we discuss a strategy to visualize the learned features from the proposed deep learning model. The cross-validation results show that our best classifier achieved an area under the curve of 0.68, which improves significantly over current clinical methods (0.58), demonstrating the potential benefit of using advanced machine learning methods in TSS classification. PMID:29854156

Age and gender classification in the wild with unsupervised feature learning

NASA Astrophysics Data System (ADS)

Wan, Lihong; Huo, Hong; Fang, Tao

2017-03-01

Inspired by unsupervised feature learning (UFL) within the self-taught learning framework, we propose a method based on UFL, convolution representation, and part-based dimensionality reduction to handle facial age and gender classification, which are two challenging problems under unconstrained circumstances. First, UFL is introduced to learn selective receptive fields (filters) automatically by applying whitening transformation and spherical k-means on random patches collected from unlabeled data. The learning process is fast and has no hyperparameters to tune. Then, the input image is convolved with these filters to obtain filtering responses on which local contrast normalization is applied. Average pooling and feature concatenation are then used to form global face representation. Finally, linear discriminant analysis with part-based strategy is presented to reduce the dimensions of the global representation and to improve classification performances further. Experiments on three challenging databases, namely, Labeled faces in the wild, Gallagher group photos, and Adience, demonstrate the effectiveness of the proposed method relative to that of state-of-the-art approaches.

Comparing the performance of flat and hierarchical Habitat/Land-Cover classification models in a NATURA 2000 site

NASA Astrophysics Data System (ADS)

Gavish, Yoni; O'Connell, Jerome; Marsh, Charles J.; Tarantino, Cristina; Blonda, Palma; Tomaselli, Valeria; Kunin, William E.

2018-02-01

The increasing need for high quality Habitat/Land-Cover (H/LC) maps has triggered considerable research into novel machine-learning based classification models. In many cases, H/LC classes follow pre-defined hierarchical classification schemes (e.g., CORINE), in which fine H/LC categories are thematically nested within more general categories. However, none of the existing machine-learning algorithms account for this pre-defined hierarchical structure. Here we introduce a novel Random Forest (RF) based application of hierarchical classification, which fits a separate local classification model in every branching point of the thematic tree, and then integrates all the different local models to a single global prediction. We applied the hierarchal RF approach in a NATURA 2000 site in Italy, using two land-cover (CORINE, FAO-LCCS) and one habitat classification scheme (EUNIS) that differ from one another in the shape of the class hierarchy. For all 3 classification schemes, both the hierarchical model and a flat model alternative provided accurate predictions, with kappa values mostly above 0.9 (despite using only 2.2-3.2% of the study area as training cells). The flat approach slightly outperformed the hierarchical models when the hierarchy was relatively simple, while the hierarchical model worked better under more complex thematic hierarchies. Most misclassifications came from habitat pairs that are thematically distant yet spectrally similar. In 2 out of 3 classification schemes, the additional constraints of the hierarchical model resulted with fewer such serious misclassifications relative to the flat model. The hierarchical model also provided valuable information on variable importance which can shed light into "black-box" based machine learning algorithms like RF. We suggest various ways by which hierarchical classification models can increase the accuracy and interpretability of H/LC classification maps.

Comparison promotes learning and transfer of relational categories.

PubMed

Kurtz, Kenneth J; Boukrina, Olga; Gentner, Dedre

2013-07-01

We investigated the effect of co-presenting training items during supervised classification learning of novel relational categories. Strong evidence exists that comparison induces a structural alignment process that renders common relational structure more salient. We hypothesized that comparisons between exemplars would facilitate learning and transfer of categories that cohere around a common relational property. The effect of comparison was investigated using learning trials that elicited a separate classification response for each item in presentation pairs that could be drawn from the same or different categories. This methodology ensures consideration of both items and invites comparison through an implicit same-different judgment inherent in making the two responses. In a test phase measuring learning and transfer, the comparison group significantly outperformed a control group receiving an equivalent training session of single-item classification learning. Comparison-based learners also outperformed the control group on a test of far transfer, that is, the ability to accurately classify items from a novel domain that was relationally alike, but surface-dissimilar, to the training materials. Theoretical and applied implications of this comparison advantage are discussed. PsycINFO Database Record (c) 2013 APA, all rights reserved.

Task-driven dictionary learning.

PubMed

Mairal, Julien; Bach, Francis; Ponce, Jean

2012-04-01

Modeling data with linear combinations of a few elements from a learned dictionary has been the focus of much recent research in machine learning, neuroscience, and signal processing. For signals such as natural images that admit such sparse representations, it is now well established that these models are well suited to restoration tasks. In this context, learning the dictionary amounts to solving a large-scale matrix factorization problem, which can be done efficiently with classical optimization tools. The same approach has also been used for learning features from data for other purposes, e.g., image classification, but tuning the dictionary in a supervised way for these tasks has proven to be more difficult. In this paper, we present a general formulation for supervised dictionary learning adapted to a wide variety of tasks, and present an efficient algorithm for solving the corresponding optimization problem. Experiments on handwritten digit classification, digital art identification, nonlinear inverse image problems, and compressed sensing demonstrate that our approach is effective in large-scale settings, and is well suited to supervised and semi-supervised classification, as well as regression tasks for data that admit sparse representations.

Application of classification-tree methods to identify nitrate sources in ground water

USGS Publications Warehouse

Spruill, T.B.; Showers, W.J.; Howe, S.S.

2002-01-01

A study was conducted to determine if nitrate sources in ground water (fertilizer on crops, fertilizer on golf courses, irrigation spray from hog (Sus scrofa) wastes, and leachate from poultry litter and septic systems) could be classified with 80% or greater success. Two statistical classification-tree models were devised from 48 water samples containing nitrate from five source categories. Model I was constructed by evaluating 32 variables and selecting four primary predictor variables (??15N, nitrate to ammonia ratio, sodium to potassium ratio, and zinc) to identify nitrate sources. A ??15N value of nitrate plus potassium 18.2 indicated inorganic or soil organic N. A nitrate to ammonia ratio 575 indicated nitrate from golf courses. A sodium to potassium ratio 3.2 indicated spray or poultry wastes. A value for zinc 2.8 indicated poultry wastes. Model 2 was devised by using all variables except ??15N. This model also included four variables (sodium plus potassium, nitrate to ammonia ratio, calcium to magnesium ratio, and sodium to potassium ratio) to distinguish categories. Both models were able to distinguish all five source categories with better than 80% overall success and with 71 to 100% success in individual categories using the learning samples. Seventeen water samples that were not used in model development were tested using Model 2 for three categories, and all were correctly classified. Classification-tree models show great potential in identifying sources of contamination and variables important in the source-identification process.

Investigating the relationship between a soils classification and the spatial parameters of a conceptual catchment-scale hydrological model

NASA Astrophysics Data System (ADS)

Dunn, S. M.; Lilly, A.

2001-10-01

There are now many examples of hydrological models that utilise the capabilities of Geographic Information Systems to generate spatially distributed predictions of behaviour. However, the spatial variability of hydrological parameters relating to distributions of soils and vegetation can be hard to establish. In this paper, the relationship between a soil hydrological classification Hydrology of Soil Types (HOST) and the spatial parameters of a conceptual catchment-scale model is investigated. A procedure involving inverse modelling using Monte-Carlo simulations on two catchments is developed to identify relative values for soil related parameters of the DIY model. The relative values determine the internal variability of hydrological processes as a function of the soil type. For three out of the four soil parameters studied, the variability between HOST classes was found to be consistent across two catchments when tested independently. Problems in identifying values for the fourth 'fast response distance' parameter have highlighted a potential limitation with the present structure of the model. The present assumption that this parameter can be related simply to soil type rather than topography appears to be inadequate. With the exclusion of this parameter, calibrated parameter sets from one catchment can be converted into equivalent parameter sets for the alternate catchment on the basis of their HOST distributions, to give a reasonable simulation of flow. Following further testing on different catchments, and modifications to the definition of the fast response distance parameter, the technique provides a methodology whereby it is possible to directly derive spatial soil parameters for new catchments.

Goal oriented soil mapping: applying modern methods supported by local knowledge: A review

NASA Astrophysics Data System (ADS)

Pereira, Paulo; Brevik, Eric; Oliva, Marc; Estebaranz, Ferran; Depellegrin, Daniel; Novara, Agata; Cerda, Artemi; Menshov, Oleksandr

2017-04-01

In the recent years the amount of soil data available increased importantly. This facilitated the production of better and accurate maps, important for sustainable land management (Pereira et al., 2017). Despite these advances, the human knowledge is extremely important to understand the natural characteristics of the landscape. The knowledge accumulated and transmitted generation after generation is priceless, and should be considered as a valuable data source for soil mapping and modelling. The local knowledge and wisdom can complement the new advances in soil analysis. In addition, farmers are the most interested in the participation and incorporation of their knowledge in the models, since they are the end-users of the study that soil scientists produce. Integration of local community's vision and understanding about nature is assumed to be an important step to the implementation of decision maker's policies. Despite this, many challenges appear regarding the integration of local and scientific knowledge, since in some cases there is no spatial correlation between folk and scientific classifications, which may be attributed to the different cultural variables that influence local soil classification. The objective of this work is to review how modern soil methods incorporated local knowledge in their models. References Pereira, P., Brevik, E., Oliva, M., Estebaranz, F., Depellegrin, D., Novara, A., Cerda, A., Menshov, O. (2017) Goal Oriented soil mapping: applying modern methods supported by local knowledge. In: Pereira, P., Brevik, E., Munoz-Rojas, M., Miller, B. (Eds.) Soil mapping and process modelling for sustainable land use management (Elsevier Publishing House) ISBN: 9780128052006

Similarity-Dissimilarity Competition in Disjunctive Classification Tasks

PubMed Central

Mathy, Fabien; Haladjian, Harry H.; Laurent, Eric; Goldstone, Robert L.

2013-01-01

Typical disjunctive artificial classification tasks require participants to sort stimuli according to rules such as “x likes cars only when black and coupe OR white and SUV.” For categories like this, increasing the salience of the diagnostic dimensions has two simultaneous effects: increasing the distance between members of the same category and increasing the distance between members of opposite categories. Potentially, these two effects respectively hinder and facilitate classification learning, leading to competing predictions for learning. Increasing saliency may lead to members of the same category to be considered less similar, while the members of separate categories might be considered more dissimilar. This implies a similarity-dissimilarity competition between two basic classification processes. When focusing on sub-category similarity, one would expect more difficult classification when members of the same category become less similar (disregarding the increase of between-category dissimilarity); however, the between-category dissimilarity increase predicts a less difficult classification. Our categorization study suggests that participants rely more on using dissimilarities between opposite categories than finding similarities between sub-categories. We connect our results to rule- and exemplar-based classification models. The pattern of influences of within- and between-category similarities are challenging for simple single-process categorization systems based on rules or exemplars. Instead, our results suggest that either these processes should be integrated in a hybrid model, or that category learning operates by forming clusters within each category. PMID:23403979

Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2's q2-feature-classifier plugin.

PubMed

Bokulich, Nicholas A; Kaehler, Benjamin D; Rideout, Jai Ram; Dillon, Matthew; Bolyen, Evan; Knight, Rob; Huttley, Gavin A; Gregory Caporaso, J

2018-05-17

Taxonomic classification of marker-gene sequences is an important step in microbiome analysis. We present q2-feature-classifier ( https://github.com/qiime2/q2-feature-classifier ), a QIIME 2 plugin containing several novel machine-learning and alignment-based methods for taxonomy classification. We evaluated and optimized several commonly used classification methods implemented in QIIME 1 (RDP, BLAST, UCLUST, and SortMeRNA) and several new methods implemented in QIIME 2 (a scikit-learn naive Bayes machine-learning classifier, and alignment-based taxonomy consensus methods based on VSEARCH, and BLAST+) for classification of bacterial 16S rRNA and fungal ITS marker-gene amplicon sequence data. The naive-Bayes, BLAST+-based, and VSEARCH-based classifiers implemented in QIIME 2 meet or exceed the species-level accuracy of other commonly used methods designed for classification of marker gene sequences that were evaluated in this work. These evaluations, based on 19 mock communities and error-free sequence simulations, including classification of simulated "novel" marker-gene sequences, are available in our extensible benchmarking framework, tax-credit ( https://github.com/caporaso-lab/tax-credit-data ). Our results illustrate the importance of parameter tuning for optimizing classifier performance, and we make recommendations regarding parameter choices for these classifiers under a range of standard operating conditions. q2-feature-classifier and tax-credit are both free, open-source, BSD-licensed packages available on GitHub.

Label-aligned Multi-task Feature Learning for Multimodal Classification of Alzheimer’s Disease and Mild Cognitive Impairment

PubMed Central

Zu, Chen; Jie, Biao; Liu, Mingxia; Chen, Songcan

2015-01-01

Multimodal classification methods using different modalities of imaging and non-imaging data have recently shown great advantages over traditional single-modality-based ones for diagnosis and prognosis of Alzheimer’s disease (AD), as well as its prodromal stage, i.e., mild cognitive impairment (MCI). However, to the best of our knowledge, most existing methods focus on mining the relationship across multiple modalities of the same subjects, while ignoring the potentially useful relationship across different subjects. Accordingly, in this paper, we propose a novel learning method for multimodal classification of AD/MCI, by fully exploring the relationships across both modalities and subjects. Specifically, our proposed method includes two subsequent components, i.e., label-aligned multi-task feature selection and multimodal classification. In the first step, the feature selection learning from multiple modalities are treated as different learning tasks and a group sparsity regularizer is imposed to jointly select a subset of relevant features. Furthermore, to utilize the discriminative information among labeled subjects, a new label-aligned regularization term is added into the objective function of standard multi-task feature selection, where label-alignment means that all multi-modality subjects with the same class labels should be closer in the new feature-reduced space. In the second step, a multi-kernel support vector machine (SVM) is adopted to fuse the selected features from multi-modality data for final classification. To validate our method, we perform experiments on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database using baseline MRI and FDG-PET imaging data. The experimental results demonstrate that our proposed method achieves better classification performance compared with several state-of-the-art methods for multimodal classification of AD/MCI. PMID:26572145

Photometric classification of type Ia supernovae in the SuperNova Legacy Survey with supervised learning

DOE Office of Scientific and Technical Information (OSTI.GOV)

Möller, A.; Ruhlmann-Kleider, V.; Leloup, C.

In the era of large astronomical surveys, photometric classification of supernovae (SNe) has become an important research field due to limited spectroscopic resources for candidate follow-up and classification. In this work, we present a method to photometrically classify type Ia supernovae based on machine learning with redshifts that are derived from the SN light-curves. This method is implemented on real data from the SNLS deferred pipeline, a purely photometric pipeline that identifies SNe Ia at high-redshifts (0.2 < z < 1.1). Our method consists of two stages: feature extraction (obtaining the SN redshift from photometry and estimating light-curve shape parameters)more » and machine learning classification. We study the performance of different algorithms such as Random Forest and Boosted Decision Trees. We evaluate the performance using SN simulations and real data from the first 3 years of the Supernova Legacy Survey (SNLS), which contains large spectroscopically and photometrically classified type Ia samples. Using the Area Under the Curve (AUC) metric, where perfect classification is given by 1, we find that our best-performing classifier (Extreme Gradient Boosting Decision Tree) has an AUC of 0.98.We show that it is possible to obtain a large photometrically selected type Ia SN sample with an estimated contamination of less than 5%. When applied to data from the first three years of SNLS, we obtain 529 events. We investigate the differences between classifying simulated SNe, and real SN survey data. In particular, we find that applying a thorough set of selection cuts to the SN sample is essential for good classification. This work demonstrates for the first time the feasibility of machine learning classification in a high- z SN survey with application to real SN data.« less

Soil organic matter degradation and enzymatic profiles of intertidal and subaqueous soils

NASA Astrophysics Data System (ADS)

Ferronato, Chiara; Marinari, Sara; Bello, Diana; Vianello, Gilmo; Trasar-Cepeda, Carmen; Vittori Antisari, Livia

2017-04-01

The interest on intertidal and subaqueous soils has recently arisen because of the climate changes forecasts. The preservation of these habitats represents an important challenge for the future of humanity, because these systems represent an important global C sink since soil organic matter (SOM) on intertidal and subaqueous soils undergoes very slow degradation rates due to oxygen limitation. Publications on SOM cycle in saltmarshes are very scarce because of the difficulties involved on those studies i.e. the interaction of many abiotic and biotic factors (e.g., redox changes, water and bio-turbation processes, etc) and stressors (e.g., salinity and anoxia). However, saltmarshes constitute an unique natural system to observe the influence of anoxic conditions on SOM degradation, because the tide fluctuations on the soil surface allow the formation of provisionally or permanently submerged soils. With the aim to investigate the quality of SOM in subaqueous soils, triplicates of subaqueous soils (SASs), intertidal soils (ITSs) and terrestrial soils (TESs) were collected in the saltmarshes of the Baiona Lagoon (Northern Italy) and classified according to their pedogenetic horizons. The SOM quality on each soil horizon was investigated by quantifying SOM, total and water-soluble organic carbon (TOC, WSC) and microbial biomass carbon (MBC). Given the contribution of soil enzymes to the degradation of SOM, some enzymatic assays were also performed. Thereafter, soil classification and humus morpho-functional classification were used to join together similar soil profiles to facilitate the description and discussion of results. Soils were ranked as Aquent or Wassent Entisols, with an A/AC/C pedosequence. SOM, TOC and MBC were statistically higher in A than in AC and C horizons. Among the A horizons, ITSs were those showing the highest values for these parameters (11% TOC, 1.6 mg kg-1 MBC, 0.9 mg kg-1 WSC). These results, combined with the morpho-functional classification of epipedons, reflect the influence of the type of annual biomass depositions on ITSs (i.e. Salicornia europaea), but also the important role of the tide oscillation that promotes the continuous alternation of red-ox exchanges and thus fasten the organic matter turnover in ITSs. On these pedons, invertase was the most effective enzymes (11.6 μmol glucose g-1h-1). Moreover, in SASs and ITSs, most of the activities linked to the degradation of exoskeletons and fungi (e.g. chitinase) increase along the soil profile, probably due to the disrupting effect of water on the soil and to the type of SOM in saltmarshes soils. By considering the specific activity (enzymatic activity/TOC content), data showed how SASs, ITSs and TESs had different oxidoreductases and hydrolases trends, suggesting a different path and effectiveness of SOM degradation, which probably depends both on the soil hydric regime, and on the different type of organic compounds. A particular increase of catalase and invertase specific activities along the soil profiles, suggests the presence of microaerophilic environment in some saturated AC and C sandy horizons but generally, it was observed a gradual decrease of biochemical alteration of the SOM by enzymatic activities along the soil profile due to the progressive restriction of the edaphic conditions.

The learning curve for narrow-band imaging in the diagnosis of precancerous gastric lesions by using Web-based video.

PubMed

Dias-Silva, Diogo; Pimentel-Nunes, Pedro; Magalhães, Joana; Magalhães, Ricardo; Veloso, Nuno; Ferreira, Carlos; Figueiredo, Pedro; Moutinho, Pedro; Dinis-Ribeiro, Mário

2014-06-01

A simplified narrow-band imaging (NBI) endoscopy classification of gastric precancerous and cancerous lesions was derived and validated in a multicenter study. This classification comes with the need for dissemination through adequate training. To address the learning curve of this classification by endoscopists with differing expertise and to assess the feasibility of a YouTube-based learning program to disseminate it. Prospective study. Five centers. Six gastroenterologists (3 trainees, 3 fully trained endoscopists [FTs]). Twenty tests provided through a Web-based program containing 10 randomly ordered NBI videos of gastric mucosa were taken. Feedback was sent 7 days after every test submission. Measures of accuracy of the NBI classification throughout the time. From the first to the last 50 videos, a learning curve was observed with a 10% increase in global accuracy, for both trainees (from 64% to 74%) and FTs (from 56% to 65%). After 200 videos, sensitivity and specificity of 80% and higher for intestinal metaplasia were observed in half the participants, and a specificity for dysplasia greater than 95%, along with a relevant likelihood ratio for a positive result of 7 to 28 and likelihood ratio for a negative result of 0.21 to 0.82, were achieved by all of the participants. No constant learning curve was observed for the identification of Helicobacter pylori gastritis and sensitivity to dysplasia. The trainees had better results in all of the parameters, except specificity for dysplasia, compared with the FTs. Globally, participants agreed that the program's structure was adequate, except on the feedback, which should have consisted of a more detailed explanation of each answer. No formal sample size estimate. A Web-based learning program could be used to teach and disseminate classifications in the endoscopy field. In this study, an NBI classification for gastric mucosal features seems to be easily learned for the identification of gastric preneoplastic lesions. Copyright © 2014 American Society for Gastrointestinal Endoscopy. Published by Mosby, Inc. All rights reserved.

Applying machine learning classification techniques to automate sky object cataloguing

NASA Astrophysics Data System (ADS)

Fayyad, Usama M.; Doyle, Richard J.; Weir, W. Nick; Djorgovski, Stanislav

1993-08-01

We describe the application of an Artificial Intelligence machine learning techniques to the development of an automated tool for the reduction of a large scientific data set. The 2nd Mt. Palomar Northern Sky Survey is nearly completed. This survey provides comprehensive coverage of the northern celestial hemisphere in the form of photographic plates. The plates are being transformed into digitized images whose quality will probably not be surpassed in the next ten to twenty years. The images are expected to contain on the order of 107 galaxies and 108 stars. Astronomers wish to determine which of these sky objects belong to various classes of galaxies and stars. Unfortunately, the size of this data set precludes analysis in an exclusively manual fashion. Our approach is to develop a software system which integrates the functions of independently developed techniques for image processing and data classification. Digitized sky images are passed through image processing routines to identify sky objects and to extract a set of features for each object. These routines are used to help select a useful set of attributes for classifying sky objects. Then GID3 (Generalized ID3) and O-B Tree, two inductive learning techniques, learns classification decision trees from examples. These classifiers will then be applied to new data. These developmnent process is highly interactive, with astronomer input playing a vital role. Astronomers refine the feature set used to construct sky object descriptions, and evaluate the performance of the automated classification technique on new data. This paper gives an overview of the machine learning techniques with an emphasis on their general applicability, describes the details of our specific application, and reports the initial encouraging results. The results indicate that our machine learning approach is well-suited to the problem. The primary benefit of the approach is increased data reduction throughput. Another benefit is consistency of classification. The classification rules which are the product of the inductive learning techniques will form an objective, examinable basis for classifying sky objects. A final, not to be underestimated benefit is that astronomers will be freed from the tedium of an intensely visual task to pursue more challenging analysis and interpretation problems based on automatically catalogued data.

Automated Grading of Gliomas using Deep Learning in Digital Pathology Images: A modular approach with ensemble of convolutional neural networks.

PubMed

Ertosun, Mehmet Günhan; Rubin, Daniel L

2015-01-01

Brain glioma is the most common primary malignant brain tumors in adults with different pathologic subtypes: Lower Grade Glioma (LGG) Grade II, Lower Grade Glioma (LGG) Grade III, and Glioblastoma Multiforme (GBM) Grade IV. The survival and treatment options are highly dependent of this glioma grade. We propose a deep learning-based, modular classification pipeline for automated grading of gliomas using digital pathology images. Whole tissue digitized images of pathology slides obtained from The Cancer Genome Atlas (TCGA) were used to train our deep learning modules. Our modular pipeline provides diagnostic quality statistics, such as precision, sensitivity and specificity, of the individual deep learning modules, and (1) facilitates training given the limited data in this domain, (2) enables exploration of different deep learning structures for each module, (3) leads to developing less complex modules that are simpler to analyze, and (4) provides flexibility, permitting use of single modules within the framework or use of other modeling or machine learning applications, such as probabilistic graphical models or support vector machines. Our modular approach helps us meet the requirements of minimum accuracy levels that are demanded by the context of different decision points within a multi-class classification scheme. Convolutional Neural Networks are trained for each module for each sub-task with more than 90% classification accuracies on validation data set, and achieved classification accuracy of 96% for the task of GBM vs LGG classification, 71% for further identifying the grade of LGG into Grade II or Grade III on independent data set coming from new patients from the multi-institutional repository.

Automated Grading of Gliomas using Deep Learning in Digital Pathology Images: A modular approach with ensemble of convolutional neural networks

PubMed Central

Ertosun, Mehmet Günhan; Rubin, Daniel L.

2015-01-01

Brain glioma is the most common primary malignant brain tumors in adults with different pathologic subtypes: Lower Grade Glioma (LGG) Grade II, Lower Grade Glioma (LGG) Grade III, and Glioblastoma Multiforme (GBM) Grade IV. The survival and treatment options are highly dependent of this glioma grade. We propose a deep learning-based, modular classification pipeline for automated grading of gliomas using digital pathology images. Whole tissue digitized images of pathology slides obtained from The Cancer Genome Atlas (TCGA) were used to train our deep learning modules. Our modular pipeline provides diagnostic quality statistics, such as precision, sensitivity and specificity, of the individual deep learning modules, and (1) facilitates training given the limited data in this domain, (2) enables exploration of different deep learning structures for each module, (3) leads to developing less complex modules that are simpler to analyze, and (4) provides flexibility, permitting use of single modules within the framework or use of other modeling or machine learning applications, such as probabilistic graphical models or support vector machines. Our modular approach helps us meet the requirements of minimum accuracy levels that are demanded by the context of different decision points within a multi-class classification scheme. Convolutional Neural Networks are trained for each module for each sub-task with more than 90% classification accuracies on validation data set, and achieved classification accuracy of 96% for the task of GBM vs LGG classification, 71% for further identifying the grade of LGG into Grade II or Grade III on independent data set coming from new patients from the multi-institutional repository. PMID:26958289

An integrative machine learning strategy for improved prediction of essential genes in Escherichia coli metabolism using flux-coupled features.

PubMed

Nandi, Sutanu; Subramanian, Abhishek; Sarkar, Ram Rup

2017-07-25

Prediction of essential genes helps to identify a minimal set of genes that are absolutely required for the appropriate functioning and survival of a cell. The available machine learning techniques for essential gene prediction have inherent problems, like imbalanced provision of training datasets, biased choice of the best model for a given balanced dataset, choice of a complex machine learning algorithm, and data-based automated selection of biologically relevant features for classification. Here, we propose a simple support vector machine-based learning strategy for the prediction of essential genes in Escherichia coli K-12 MG1655 metabolism that integrates a non-conventional combination of an appropriate sample balanced training set, a unique organism-specific genotype, phenotype attributes that characterize essential genes, and optimal parameters of the learning algorithm to generate the best machine learning model (the model with the highest accuracy among all the models trained for different sample training sets). For the first time, we also introduce flux-coupled metabolic subnetwork-based features for enhancing the classification performance. Our strategy proves to be superior as compared to previous SVM-based strategies in obtaining a biologically relevant classification of genes with high sensitivity and specificity. This methodology was also trained with datasets of other recent supervised classification techniques for essential gene classification and tested using reported test datasets. The testing accuracy was always high as compared to the known techniques, proving that our method outperforms known methods. Observations from our study indicate that essential genes are conserved among homologous bacterial species, demonstrate high codon usage bias, GC content and gene expression, and predominantly possess a tendency to form physiological flux modules in metabolism.

Deep Learning in Label-free Cell Classification

DOE Office of Scientific and Technical Information (OSTI.GOV)

Chen, Claire Lifan; Mahjoubfar, Ata; Tai, Li-Chia

Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individualmore » cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. In conclusion, this system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.« less

Case-based statistical learning applied to SPECT image classification

NASA Astrophysics Data System (ADS)

Górriz, Juan M.; Ramírez, Javier; Illán, I. A.; Martínez-Murcia, Francisco J.; Segovia, Fermín.; Salas-Gonzalez, Diego; Ortiz, A.

2017-03-01

Statistical learning and decision theory play a key role in many areas of science and engineering. Some examples include time series regression and prediction, optical character recognition, signal detection in communications or biomedical applications for diagnosis and prognosis. This paper deals with the topic of learning from biomedical image data in the classification problem. In a typical scenario we have a training set that is employed to fit a prediction model or learner and a testing set on which the learner is applied to in order to predict the outcome for new unseen patterns. Both processes are usually completely separated to avoid over-fitting and due to the fact that, in practice, the unseen new objects (testing set) have unknown outcomes. However, the outcome yields one of a discrete set of values, i.e. the binary diagnosis problem. Thus, assumptions on these outcome values could be established to obtain the most likely prediction model at the training stage, that could improve the overall classification accuracy on the testing set, or keep its performance at least at the level of the selected statistical classifier. In this sense, a novel case-based learning (c-learning) procedure is proposed which combines hypothesis testing from a discrete set of expected outcomes and a cross-validated classification stage.

Deep Learning in Label-free Cell Classification

DOE PAGES

Chen, Claire Lifan; Mahjoubfar, Ata; Tai, Li-Chia; ...

2016-03-15

Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individualmore » cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. In conclusion, this system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.« less

Histopathological Image Classification using Discriminative Feature-oriented Dictionary Learning

PubMed Central

Vu, Tiep Huu; Mousavi, Hojjat Seyed; Monga, Vishal; Rao, Ganesh; Rao, UK Arvind

2016-01-01

In histopathological image analysis, feature extraction for classification is a challenging task due to the diversity of histology features suitable for each problem as well as presence of rich geometrical structures. In this paper, we propose an automatic feature discovery framework via learning class-specific dictionaries and present a low-complexity method for classification and disease grading in histopathology. Essentially, our Discriminative Feature-oriented Dictionary Learning (DFDL) method learns class-specific dictionaries such that under a sparsity constraint, the learned dictionaries allow representing a new image sample parsimoniously via the dictionary corresponding to the class identity of the sample. At the same time, the dictionary is designed to be poorly capable of representing samples from other classes. Experiments on three challenging real-world image databases: 1) histopathological images of intraductal breast lesions, 2) mammalian kidney, lung and spleen images provided by the Animal Diagnostics Lab (ADL) at Pennsylvania State University, and 3) brain tumor images from The Cancer Genome Atlas (TCGA) database, reveal the merits of our proposal over state-of-the-art alternatives. Moreover, we demonstrate that DFDL exhibits a more graceful decay in classification accuracy against the number of training images which is highly desirable in practice where generous training is often not available. PMID:26513781

Methodological Issues in Predicting Pediatric Epilepsy Surgery Candidates Through Natural Language Processing and Machine Learning

PubMed Central

Cohen, Kevin Bretonnel; Glass, Benjamin; Greiner, Hansel M.; Holland-Bouley, Katherine; Standridge, Shannon; Arya, Ravindra; Faist, Robert; Morita, Diego; Mangano, Francesco; Connolly, Brian; Glauser, Tracy; Pestian, John

2016-01-01

Objective: We describe the development and evaluation of a system that uses machine learning and natural language processing techniques to identify potential candidates for surgical intervention for drug-resistant pediatric epilepsy. The data are comprised of free-text clinical notes extracted from the electronic health record (EHR). Both known clinical outcomes from the EHR and manual chart annotations provide gold standards for the patient’s status. The following hypotheses are then tested: 1) machine learning methods can identify epilepsy surgery candidates as well as physicians do and 2) machine learning methods can identify candidates earlier than physicians do. These hypotheses are tested by systematically evaluating the effects of the data source, amount of training data, class balance, classification algorithm, and feature set on classifier performance. The results support both hypotheses, with F-measures ranging from 0.71 to 0.82. The feature set, classification algorithm, amount of training data, class balance, and gold standard all significantly affected classification performance. It was further observed that classification performance was better than the highest agreement between two annotators, even at one year before documented surgery referral. The results demonstrate that such machine learning methods can contribute to predicting pediatric epilepsy surgery candidates and reducing lag time to surgery referral. PMID:27257386

Classification of L2 Vocabulary Learning Strategies: Evidence from Exploratory and Confirmatory Factor Analyses

ERIC Educational Resources Information Center

Zhang, Bo; Li, Changyu

2011-01-01

This research presents a classification theory for the L2 vocabulary learning strategies. Based on the exploratory and confirmatory factor analyses of strategies that adult Chinese English learners used, this theory identifies six categories, four of which are related to the cognitive process in lexical acquisition and the other two are…

What Belongs in Your 15-Bean Soup? Using the Learning Cycle to Address Misconceptions about Construction of Taxonomic Keys

ERIC Educational Resources Information Center

Ross, Ann; Vanderspool, Staria

2004-01-01

Students can use seed characteristics to discriminate between the different kinds of legumes using taxonomic classification processes of sorting and ranking, followed by construction of taxonomic keys. The application of the Learning Cycle process to taxonomic principles, hierarchical classification, and construction of keys presents the…

Semantic Linking of Learning Object Repositories to DBpedia

ERIC Educational Resources Information Center

Lama, Manuel; Vidal, Juan C.; Otero-Garcia, Estefania; Bugarin, Alberto; Barro, Senen

2012-01-01

Large-sized repositories of learning objects (LOs) are difficult to create and also to maintain. In this paper we propose a way to reduce this drawback by improving the classification mechanisms of the LO repositories. Specifically, we present a solution to automate the LO classification of the Universia repository, a collection of more than 15…

Classification Agreement Analysis of Cross-Battery Assessment in the Identification of Specific Learning Disorders in Children and Youth

ERIC Educational Resources Information Center

Kranzler, John H.; Floyd, Randy G.; Benson, Nicholas; Zaboski, Brian; Thibodaux, Lia

2016-01-01

The Cross-Battery Assessment (XBA) approach to identifying a specific learning disorder (SLD) is based on the postulate that deficits in cognitive abilities in the presence of otherwise average general intelligence are causally related to academic achievement weaknesses. To examine this postulate, we conducted a classification agreement analysis…

Classifying Black Hole States with Machine Learning

NASA Astrophysics Data System (ADS)

Huppenkothen, Daniela

2018-01-01

Galactic black hole binaries are known to go through different states with apparent signatures in both X-ray light curves and spectra, leading to important implications for accretion physics as well as our knowledge of General Relativity. Existing frameworks of classification are usually based on human interpretation of low-dimensional representations of the data, and generally only apply to fairly small data sets. Machine learning, in contrast, allows for rapid classification of large, high-dimensional data sets. In this talk, I will report on advances made in classification of states observed in Black Hole X-ray Binaries, focusing on the two sources GRS 1915+105 and Cygnus X-1, and show both the successes and limitations of using machine learning to derive physical constraints on these systems.

Aggregative Learning Method and Its Application for Communication Quality Evaluation

NASA Astrophysics Data System (ADS)

Akhmetov, Dauren F.; Kotaki, Minoru

2007-12-01

In this paper, so-called Aggregative Learning Method (ALM) is proposed to improve and simplify the learning and classification abilities of different data processing systems. It provides a universal basis for design and analysis of mathematical models of wide class. A procedure was elaborated for time series model reconstruction and analysis for linear and nonlinear cases. Data approximation accuracy (during learning phase) and data classification quality (during recall phase) are estimated from introduced statistic parameters. The validity and efficiency of the proposed approach have been demonstrated through its application for monitoring of wireless communication quality, namely, for Fixed Wireless Access (FWA) system. Low memory and computation resources were shown to be needed for the procedure realization, especially for data classification (recall) stage. Characterized with high computational efficiency and simple decision making procedure, the derived approaches can be useful for simple and reliable real-time surveillance and control system design.

Tensorial dynamic time warping with articulation index representation for efficient audio-template learning.

PubMed

Le, Long N; Jones, Douglas L

2018-03-01

Audio classification techniques often depend on the availability of a large labeled training dataset for successful performance. However, in many application domains of audio classification (e.g., wildlife monitoring), obtaining labeled data is still a costly and laborious process. Motivated by this observation, a technique is proposed to efficiently learn a clean template from a few labeled, but likely corrupted (by noise and interferences), data samples. This learning can be done efficiently via tensorial dynamic time warping on the articulation index-based time-frequency representations of audio data. The learned template can then be used in audio classification following the standard template-based approach. Experimental results show that the proposed approach outperforms both (1) the recurrent neural network approach and (2) the state-of-the-art in the template-based approach on a wildlife detection application with few training samples.

Unsupervised Feature Learning for Heart Sounds Classification Using Autoencoder

NASA Astrophysics Data System (ADS)

Hu, Wei; Lv, Jiancheng; Liu, Dongbo; Chen, Yao

2018-04-01

Cardiovascular disease seriously threatens the health of many people. It is usually diagnosed during cardiac auscultation, which is a fast and efficient method of cardiovascular disease diagnosis. In recent years, deep learning approach using unsupervised learning has made significant breakthroughs in many fields. However, to our knowledge, deep learning has not yet been used for heart sound classification. In this paper, we first use the average Shannon energy to extract the envelope of the heart sounds, then find the highest point of S1 to extract the cardiac cycle. We convert the time-domain signals of the cardiac cycle into spectrograms and apply principal component analysis whitening to reduce the dimensionality of the spectrogram. Finally, we apply a two-layer autoencoder to extract the features of the spectrogram. The experimental results demonstrate that the features from the autoencoder are suitable for heart sound classification.

Performances of Machine Learning Algorithms for Binary Classification of Network Anomaly Detection System

NASA Astrophysics Data System (ADS)

Nawir, Mukrimah; Amir, Amiza; Lynn, Ong Bi; Yaakob, Naimah; Badlishah Ahmad, R.

2018-05-01

The rapid growth of technologies might endanger them to various network attacks due to the nature of data which are frequently exchange their data through Internet and large-scale data that need to be handle. Moreover, network anomaly detection using machine learning faced difficulty when dealing the involvement of dataset where the number of labelled network dataset is very few in public and this caused many researchers keep used the most commonly network dataset (KDDCup99) which is not relevant to employ the machine learning (ML) algorithms for a classification. Several issues regarding these available labelled network datasets are discussed in this paper. The aim of this paper to build a network anomaly detection system using machine learning algorithms that are efficient, effective and fast processing. The finding showed that AODE algorithm is performed well in term of accuracy and processing time for binary classification towards UNSW-NB15 dataset.

Convex formulation of multiple instance learning from positive and unlabeled bags.

PubMed

Bao, Han; Sakai, Tomoya; Sato, Issei; Sugiyama, Masashi

2018-05-24

Multiple instance learning (MIL) is a variation of traditional supervised learning problems where data (referred to as bags) are composed of sub-elements (referred to as instances) and only bag labels are available. MIL has a variety of applications such as content-based image retrieval, text categorization, and medical diagnosis. Most of the previous work for MIL assume that training bags are fully labeled. However, it is often difficult to obtain an enough number of labeled bags in practical situations, while many unlabeled bags are available. A learning framework called PU classification (positive and unlabeled classification) can address this problem. In this paper, we propose a convex PU classification method to solve an MIL problem. We experimentally show that the proposed method achieves better performance with significantly lower computation costs than an existing method for PU-MIL. Copyright © 2018 Elsevier Ltd. All rights reserved.

A Fly-Inspired Mushroom Bodies Model for Sensory-Motor Control Through Sequence and Subsequence Learning.

PubMed

Arena, Paolo; Calí, Marco; Patané, Luca; Portera, Agnese; Strauss, Roland

2016-09-01

Classification and sequence learning are relevant capabilities used by living beings to extract complex information from the environment for behavioral control. The insect world is full of examples where the presentation time of specific stimuli shapes the behavioral response. On the basis of previously developed neural models, inspired by Drosophila melanogaster, a new architecture for classification and sequence learning is here presented under the perspective of the Neural Reuse theory. Classification of relevant input stimuli is performed through resonant neurons, activated by the complex dynamics generated in a lattice of recurrent spiking neurons modeling the insect Mushroom Bodies neuropile. The network devoted to context formation is able to reconstruct the learned sequence and also to trace the subsequences present in the provided input. A sensitivity analysis to parameter variation and noise is reported. Experiments on a roving robot are reported to show the capabilities of the architecture used as a neural controller.

A Problem-Based Learning Approach to Teaching Introductory Soil Science

ERIC Educational Resources Information Center

Amador, Jose A.; Gorres, Josef H.

2004-01-01

At most land-grant universities in the USA, Introduction to Soil Science is traditionally taught using a combination of lecture and laboratory formats. To promote engagement, improve comprehension, and enhance retention of content by students, we developed a problem-based learning (PBL) introductory soil science course. Students work in groups to…

A Mixtures-of-Trees Framework for Multi-Label Classification

PubMed Central

Hong, Charmgil; Batal, Iyad; Hauskrecht, Milos

2015-01-01

We propose a new probabilistic approach for multi-label classification that aims to represent the class posterior distribution P(Y|X). Our approach uses a mixture of tree-structured Bayesian networks, which can leverage the computational advantages of conditional tree-structured models and the abilities of mixtures to compensate for tree-structured restrictions. We develop algorithms for learning the model from data and for performing multi-label predictions using the learned model. Experiments on multiple datasets demonstrate that our approach outperforms several state-of-the-art multi-label classification methods. PMID:25927011

Data Processing And Machine Learning Methods For Multi-Modal Operator State Classification Systems

NASA Technical Reports Server (NTRS)

Hearn, Tristan A.

2015-01-01

This document is intended as an introduction to a set of common signal processing learning methods that may be used in the software portion of a functional crew state monitoring system. This includes overviews of both the theory of the methods involved, as well as examples of implementation. Practical considerations are discussed for implementing modular, flexible, and scalable processing and classification software for a multi-modal, multi-channel monitoring system. Example source code is also given for all of the discussed processing and classification methods.

Mapping of topsoil organic carbon in agro-ecosystems of a flat terrain area (Lombardy) by means of legacy soil data, climatic data and NDVI time series predictors with machine learning methods

NASA Astrophysics Data System (ADS)

Schillaci, Calogero; Saia, Sergio; Braun, Andreas; Acutis, Marco

2017-04-01

Topsoil organic carbon plays an important role in the agricultural yield, yield potential, and to deliver many ecosystem services, such as the potential to reduce greenhouse gas (GHG) emission from soil. In particular, SOC content sturdily affects soil properties, thus the precision of its estimation can support broad-scale agricultural and environmental management policy. Soils in temperate agro-ecosystem are generally highly productive and cropland occupies about 60% of their surface (Ramankutty et al 2008). In such contexts, lands is frequently subjected to SOC degrading operations, mostly ploughing, with drawbacks on soil fertility and erosion. In temperate agro-ecosystems, a strong role in SOC maintenance can be played by manure and residues inputs after husbandry and related activities and return of plant biomass to the soil (Acutis et al 2014). In this perspective, soil management can have a major role in SOC spatial distribution to maintain soil fertility and ecosystem services in a target area. Due to the considerable importance of SOC on both agronomical and ecological aspects of the agro-ecosystems, regional soil surveys over the years frequently take into account the measurement of SOC concentration and often stock. In the present study, we integrated a highly detailed legacy SOC dataset with climatic data and RS data to produce a reliable SOC maps from a farm to a district scale. In particular, the Normalized Difference Vegetation Index (NDVI)was used after the computation of its average value in a given pixel derived from several approximately cloud-free images. The input dataset was made of about 3000 Ap horizons implemented of SOC concentration, texture, bulk density and metadata. Climatic data (Worldclim), soil type (from the pedological map 1:250000 WRB), and a time series NDVI were applied. The NDVI data were derived from a set of Landsat 5 scenes (path 193, row 28,29) whereas the path 194, (row 28 and 29) contributes for less than one fourth of the study area. The use of machine learning approach for the generation of a SOC map of the flat terrain agricultural topsoil of Lombardy using the regional soil database relies on two assumptions: (1) the slow change in the content of the stabilised soil organic matter (SOM) fraction, which is almost everywhere the most represented SOM fraction; and (2) the intrinsic low erosion rates due to the low mean slope. In particular, NDVI, which is related land cover and to the amount of biomass returned the soil, can have drawbacks when applied in cultivated fields. These drawbacks mainly concern the variability on crop biomass within and across the year. Notwithstanding, this issue makes NDVI very suitable for differentiating contrasting land use (e.g. field crops vs. orchards) when computed from images captured in particular crop cycle moments (e.g. in summer). However, the same issue reduces NDVI suitability to estimate the amount of biomass within each land use or when aiming at highly detailed resolution. Different grade of cloud cover were admitted to construct the average NDVI. Boosted regression trees were used to reveal the effect of each spatial covariate in predicting the SOC content. Preliminary results highlighted that the integration of the soil pedological classification and the mean NDVI improved the pixel classification in SOC classes according to crop type and management. As expected, climatic gradient played an important role in SOC modelling but did not affect the spatial resolution of the final map. In conclusion, SOC estimate strongly depends on sample density and homogeneity of distribution and the environmental heterogeneity. The lack of the strong topographical traits in flat terrain areas represents a challenge for soil mapping. In such conditions, the computation of a reliable biomass-related RS trait such as the mean NDVI can increase the prediction ability of the models and reduce the mapping biases. References Acutis, M., Alfieri, L., Giussani, A., Provolo, G., Di Guardo, A., Colombini, S., Bertoncini, G.,Castelnuovo, M., Sali, G., Moschini, M., Sanna, M., Perego, A., Carozzi, M., Chiodini, M.E., Fumagalli, M., 2014. ValorE: An integrated and GIS-based decision support system for livestock manure management in the Lombardy region (northern Italy). Land use policy 41, 149-162. doi:10.1016/j.landusepol.2014.05.007 Ramankutty, N., A. T. Evan, C. Monfreda, and J. A. Foley (2008), Farming the planet: 1. Geographic distribution of global agricultural lands in the year 2000, Global Biogeochem. Cycles , 22, GB1003, doi:10.1029/2007GB002952.

Centrifugal Modelling of Soil Structures. Part I. Centrifugal Modelling of Slope Failures.

DTIC Science & Technology

1979-03-01

comparing successive photographs in which soil movement was noted by the change in position of the original grid of silvered indicator balls . Inherent in...SECIJ RITY CLASSIFICATION OF THIS PAGE(1Thon Pat& Entered) of uplift forces was also observed. In nineteen coal mine waste embankment dam models...In’nineteen coal mine waste embankment dam models, throughout which the soil particle size distribution was altered for modelling of dif- ferent

Development of SMAP Mission Cal/Val Activities

NASA Technical Reports Server (NTRS)

Colliander, A.; Jackson, T.; Kimball, J.; Cosh, M.; Spencer, M.; Entekhabi, D.; Njoku, E.; ONeill, P.

2010-01-01

The Soil Moisture Active Passive (SMAP) mission is a NASA directed mission to map global land surface soil moisture and freeze-thaw state. Instrument and mission details are shown. The key SMAP soil moisture product is provided at 10 km resolution with 0.04cubic cm/cubic cm accuracy. The freeze/thaw product is provided at 3 km resolution and 80% frozen-thawed classification accuracy. The full list of SMAP data products is shown.

Transfer learning improves supervised image segmentation across imaging protocols.

PubMed

van Opbroek, Annegreet; Ikram, M Arfan; Vernooij, Meike W; de Bruijne, Marleen

2015-05-01

The variation between images obtained with different scanners or different imaging protocols presents a major challenge in automatic segmentation of biomedical images. This variation especially hampers the application of otherwise successful supervised-learning techniques which, in order to perform well, often require a large amount of labeled training data that is exactly representative of the target data. We therefore propose to use transfer learning for image segmentation. Transfer-learning techniques can cope with differences in distributions between training and target data, and therefore may improve performance over supervised learning for segmentation across scanners and scan protocols. We present four transfer classifiers that can train a classification scheme with only a small amount of representative training data, in addition to a larger amount of other training data with slightly different characteristics. The performance of the four transfer classifiers was compared to that of standard supervised classification on two magnetic resonance imaging brain-segmentation tasks with multi-site data: white matter, gray matter, and cerebrospinal fluid segmentation; and white-matter-/MS-lesion segmentation. The experiments showed that when there is only a small amount of representative training data available, transfer learning can greatly outperform common supervised-learning approaches, minimizing classification errors by up to 60%.

High-resolution digital mapping of soil organic carbon in permafrost terrain using machine learning: a case study in a sub-Arctic peatland environment

NASA Astrophysics Data System (ADS)

Siewert, Matthias B.

2018-03-01

Soil organic carbon (SOC) stored in northern peatlands and permafrost-affected soils are key components in the global carbon cycle. This article quantifies SOC stocks in a sub-Arctic mountainous peatland environment in the discontinuous permafrost zone in Abisko, northern Sweden. Four machine-learning techniques are evaluated for SOC quantification: multiple linear regression, artificial neural networks, support vector machine and random forest. The random forest model performed best and was used to predict SOC for several depth increments at a spatial resolution of 1 m (1×1 m). A high-resolution (1 m) land cover classification generated for this study is the most relevant predictive variable. The landscape mean SOC storage (0-150 cm) is estimated to be 8.3 ± 8.0 kg C m-2 and the SOC stored in the top meter (0-100 cm) to be 7.7 ± 6.2 kg C m-2. The predictive modeling highlights the relative importance of wetland areas and in particular peat plateaus for the landscape's SOC storage. The total SOC was also predicted at reduced spatial resolutions of 2, 10, 30, 100, 250 and 1000 m and shows a significant drop in land cover class detail and a tendency to underestimate the SOC at resolutions > 30 m. This is associated with the occurrence of many small-scale wetlands forming local hot-spots of SOC storage that are omitted at coarse resolutions. Sharp transitions in SOC storage associated with land cover and permafrost distribution are the most challenging methodological aspect. However, in this study, at local, regional and circum-Arctic scales, the main factor limiting robust SOC mapping efforts is the scarcity of soil pedon data from across the entire environmental space. For the Abisko region, past SOC and permafrost dynamics indicate that most of the SOC is barely 2000 years old and very dynamic. Future research needs to investigate the geomorphic response of permafrost degradation and the fate of SOC across all landscape compartments in post-permafrost landscapes.

Successional stage of biological soil crusts: an accurate indicator of ecohydrological condition

USGS Publications Warehouse

Belnap, Jayne; Wilcox, Bradford P.; Van Scoyoc, Matthew V.; Phillips, Susan L.

2013-01-01

Biological soil crusts are a key component of many dryland ecosystems. Following disturbance, biological soil crusts will recover in stages. Recently, a simple classification of these stages has been developed, largely on the basis of external features of the crusts, which reflects their level of development (LOD). The classification system has six LOD classes, from low (1) to high (6). To determine whether the LOD of a crust is related to its ecohydrological function, we used rainfall simulation to evaluate differences in infiltration, runoff, and erosion among crusts in the various LODs, across a range of soil depths and with different wetting pre-treatments. We found large differences between the lowest and highest LODs, with runoff and erosion being greatest from the lowest LOD. Under dry antecedent conditions, about 50% of the water applied ran off the lowest LOD plots, whereas less than 10% ran off the plots of the two highest LODs. Similarly, sediment loss was 400 g m-2 from the lowest LOD and almost zero from the higher LODs. We scaled up the results from these simulations using the Rangeland Hydrology and Erosion Model. Modelling results indicate that erosion increases dramatically as slope length and gradient increase, especially beyond the threshold values of 10 m for slope length and 10% for slope gradient. Our findings confirm that the LOD classification is a quick, easy, nondestructive, and accurate index of hydrological condition and should be incorporated in field and modelling assessments of ecosystem health.

Soil Quality Indicator: a new concept

NASA Astrophysics Data System (ADS)

Barão, Lúcia; Basch, Gottlieb

2017-04-01

During the last century, cultivated soils have been intensively exploited for food and feed production. This exploitation has compromised the soils' natural functions and many of the soil-mediated ecosystems services, including its production potential for agriculture. Also, soils became increasingly vulnerable and less resilient to a wide range of threats. To overcome this situation, new and better management practices are needed to prevent soil from degradation. However, to adopt the best management practices in a specific location, it is necessary to evaluate the soil quality status first. Different soil quality indicators have been suggested over the last decades in order to evaluate the soil status, and those are often based on the performance of soil chemical, physical and biological properties. However, the direct link between these properties and the associated soil functions or soil vulnerability to threats appears more difficult to be established. This present work is part of the iSQAPER project- Interactive Soil Quality Assessment in Europe and China for Agricultural Productivity and Environmental Resilience, where new soil quality concepts are explored to provide better information regarding the effects of the most promising agricultural management practices on soil quality. We have developed a new conceptual soil quality indicator which determines the soil quality status, regarding its vulnerability towards different threats. First, different indicators were specifically developed for each of the eight threats considered - Erosion, SOM decline, Poor Structure, Poor water holding capacity, Compaction, N-Leaching, Soil-borne pests and diseases and Salinization. As an example for the case of Erosion, the RUSLE equation for the estimate of the soil annual loss was used. Secondly, a reference classification was established for each indicator to integrate all possible results into a Good, Intermediate or Bad classification. Finally, all indicators were combined to return a single evaluation of the soil status, using different techniques that are dependent on the final use of the soil quality indicator. Some of the advantages of this new concept include the evaluation of soil quality based on its vulnerability to threats, together with the evaluation of soil properties in a given context while also suggesting soil management practices that are directly capable to mitigate soil vulnerability towards specific threats. Keywords: Soil Quality, Agriculture, Sustainability, Soil threats

Ecological classification and management characteristics of montane forest land in southwestern Washington.

Treesearch

D.G. Brockway; C. Topik

1984-01-01

Vegetation, soil, and site data werecollectedthroughout the forested portion of the Pacific silver fir and mountain hemlock zones of the Gifford Pinchot National Forest as part of the Forest Service program to develop anecoIogicallybasedplant association classification system for the Pacific Northwest Region. The major objective of sampling was to include a wide...

Field validation of Burned Area Reflectance Classification (BARC) products for post fire assessment

Treesearch

Andrew T. Hudak; Peter R. Robichaud; Jeffery B. Evans; Jess Clark; Keith Lannom; Penelope Morgan; Carter Stone

2004-01-01

The USFS Remote Sensing Applications Center (RSAC) and the USGS EROS Data Center (EDC) produce Burned Area Reflectance Classification (BARC) maps for use by Burned Area Emergency Rehabilitation (BAER) teams in rapid response to wildfires. BAER teams desire maps indicative of soil burn severity, but photosynthetic and nonphotosynthetic vegetation also influences the...