Science.gov

Sample records for abstract machine model

  1. Abstract quantum computing machines and quantum computational logics

    NASA Astrophysics Data System (ADS)

    Chiara, Maria Luisa Dalla; Giuntini, Roberto; Sergioli, Giuseppe; Leporini, Roberto

    2016-06-01

    Classical and quantum parallelism are deeply different, although it is sometimes claimed that quantum Turing machines are nothing but special examples of classical probabilistic machines. We introduce the concepts of deterministic state machine, classical probabilistic state machine and quantum state machine. On this basis, we discuss the question: To what extent can quantum state machines be simulated by classical probabilistic state machines? Each state machine is devoted to a single task determined by its program. Real computers, however, behave differently, being able to solve different kinds of problems. This capacity can be modeled, in the quantum case, by the mathematical notion of abstract quantum computing machine, whose different programs determine different quantum state machines. The computations of abstract quantum computing machines can be linguistically described by the formulas of a particular form of quantum logic, termed quantum computational logic.

  2. Three-dimensional eddy current solution of a polyphase machine test model (abstract)

    NASA Astrophysics Data System (ADS)

    Pahner, Uwe; Belmans, Ronnie; Ostovic, Vlado

    1994-05-01

    This abstract describes a three-dimensional (3D) finite element solution of a test model that has been reported in the literature. The model is a basis for calculating the current redistribution effects in the end windings of turbogenerators. The aim of the study is to see whether the analytical results of the test model can be found using a general purpose finite element package, thus indicating that the finite element model is accurate enough to treat real end winding problems. The real end winding problems cannot be solved analytically, as the geometry is far too complicated. The model consists of a polyphase coil set, containing 44 individual coils. This set generates a two pole mmf distribution on a cylindrical surface. The rotating field causes eddy currents to flow in the inner massive and conducting rotor. In the analytical solution a perfect sinusoidal mmf distribution is put forward. The finite element model contains 85824 tetrahedra and 16451 nodes. A complex single scalar potential representation is used in the nonconducting parts. The computation time required was 3 h and 42 min. The flux plots show that the field distribution is acceptable. Furthermore, the induced currents are calculated and compared with the values found from the analytical solution. The distribution of the eddy currents is very close to the distribution of the analytical solution. The most important results are the losses, both local and global. The value of the overall losses is less than 2% away from those of the analytical solution. Also the local distribution of the losses is at any given point less than 7% away from the analytical solution. The deviations of the results are acceptable and are partially due to the fact that the sinusoidal mmf distribution was not modeled perfectly in the finite element method.

  3. Programming the Navier-Stokes computer: An abstract machine model and a visual editor

    NASA Technical Reports Server (NTRS)

    Middleton, David; Crockett, Tom; Tomboulian, Sherry

    1988-01-01

    The Navier-Stokes computer is a parallel computer designed to solve Computational Fluid Dynamics problems. Each processor contains several floating point units which can be configured under program control to implement a vector pipeline with several inputs and outputs. Since the development of an effective compiler for this computer appears to be very difficult, machine level programming seems necessary and support tools for this process have been studied. These support tools are organized into a graphical program editor. A programming process is described by which appropriate computations may be efficiently implemented on the Navier-Stokes computer. The graphical editor would support this programming process, verifying various programmer choices for correctness and deducing values such as pipeline delays and network configurations. Step by step details are provided and demonstrated with two example programs.

  4. Abstract Machines for Polymorphous Computing

    DTIC Science & Technology

    2007-12-01

    s/ /s/ MARK NOVAK WARREN H. DEBANY, Jr. Work Unit Manager Technical Advisor, Information Grid Division Information...models and LLCs have been developed for Raw, MONARCH [18][19], TRIPS [20][21], and Smart Memories [22][23]. These research projects were conducted...used here. In our approach on Raw, two key concepts are used to fully leverage the Raw architecture [34]. First, the tile grid is viewed as a

  5. Automatic Review of Abstract State Machines by Meta Property Verification

    NASA Technical Reports Server (NTRS)

    Arcaini, Paolo; Gargantini, Angelo; Riccobene, Elvinia

    2010-01-01

    A model review is a validation technique aimed at determining if a model is of sufficient quality and allows defects to be identified early in the system development, reducing the cost of fixing them. In this paper we propose a technique to perform automatic review of Abstract State Machine (ASM) formal specifications. We first detect a family of typical vulnerabilities and defects a developer can introduce during the modeling activity using the ASMs and we express such faults as the violation of meta-properties that guarantee certain quality attributes of the specification. These meta-properties are then mapped to temporal logic formulas and model checked for their violation. As a proof of concept, we also report the result of applying this ASM review process to several specifications.

  6. Modelling Metamorphism by Abstract Interpretation

    NASA Astrophysics Data System (ADS)

    Dalla Preda, Mila; Giacobazzi, Roberto; Debray, Saumya; Coogan, Kevin; Townsend, Gregg M.

    Metamorphic malware apply semantics-preserving transformations to their own code in order to foil detection systems based on signature matching. In this paper we consider the problem of automatically extract metamorphic signatures from these malware. We introduce a semantics for self-modifying code, later called phase semantics, and prove its correctness by showing that it is an abstract interpretation of the standard trace semantics. Phase semantics precisely models the metamorphic code behavior by providing a set of traces of programs which correspond to the possible evolutions of the metamorphic code during execution. We show that metamorphic signatures can be automatically extracted by abstract interpretation of the phase semantics, and that regular metamorphism can be modelled as finite state automata abstraction of the phase semantics.

  7. Machine characterization based on an abstract high-level language machine

    NASA Technical Reports Server (NTRS)

    Saavedra-Barrera, Rafael H.; Smith, Alan Jay; Miya, Eugene

    1989-01-01

    Measurements are presented for a large number of machines ranging from small workstations to supercomputers. The authors combine these measurements into groups of parameters which relate to specific aspects of the machine implementation, and use these groups to provide overall machine characterizations. The authors also define the concept of pershapes, which represent the level of performance of a machine for different types of computation. A metric based on pershapes is introduced that provides a quantitative way of measuring how similar two machines are in terms of their performance distributions. The metric is related to the extent to which pairs of machines have varying relative performance levels depending on which benchmark is used.

  8. Modeling Abstraction and Simulation Techniques

    DTIC Science & Technology

    2002-12-01

    for data reduction on the patterns stored in normal database . In [58], J. Marin et al. proposed a hybrid model to profile user behavior by the...conv(Ad) as the feasible set 124 in the “surrogate” continuous state space. When the feasible set Ad is not a polyhedron , the set Ac = conv(Ad) may...and is not necessarily convex. Note also that the definition reduces to Ac = conv(Ad) when Ad is formed by all the discrete points in a polyhedron . Now

  9. Abstraction and model evaluation in category learning.

    PubMed

    Vanpaemel, Wolf; Storms, Gert

    2010-05-01

    Thirty previously published data sets, from seminal category learning tasks, are reanalyzed using the varying abstraction model (VAM). Unlike a prototype-versus-exemplar analysis, which focuses on extreme levels of abstraction only, a VAM analysis also considers the possibility of partial abstraction. Whereas most data sets support no abstraction when only the extreme possibilities are considered, we show that evidence for abstraction can be provided using the broader view on abstraction provided by the VAM. The present results generalize earlier demonstrations of partial abstraction (Vanpaemel & Storms, 2008), in which only a small number of data sets was analyzed. Following the dominant modus operandi in category learning research, Vanpaemel and Storms evaluated the models on their best fit, a practice known to ignore the complexity of the models under consideration. In the present study, in contrast, model evaluation not only relies on the maximal likelihood, but also on the marginal likelihood, which is sensitive to model complexity. Finally, using a large recovery study, it is demonstrated that, across the 30 data sets, complexity differences between the models in the VAM family are small. This indicates that a (computationally challenging) complexity-sensitive model evaluation method is uncalled for, and that the use of a (computationally straightforward) complexity-insensitive model evaluation method is justified.

  10. Model-based machine learning

    PubMed Central

    Bishop, Christopher M.

    2013-01-01

    Several decades of research in the field of machine learning have resulted in a multitude of different algorithms for solving a broad range of problems. To tackle a new application, a researcher typically tries to map their problem onto one of these existing methods, often influenced by their familiarity with specific algorithms and by the availability of corresponding software implementations. In this study, we describe an alternative methodology for applying machine learning, in which a bespoke solution is formulated for each new application. The solution is expressed through a compact modelling language, and the corresponding custom machine learning code is then generated automatically. This model-based approach offers several major advantages, including the opportunity to create highly tailored models for specific scenarios, as well as rapid prototyping and comparison of a range of alternative models. Furthermore, newcomers to the field of machine learning do not have to learn about the huge range of traditional methods, but instead can focus their attention on understanding a single modelling environment. In this study, we show how probabilistic graphical models, coupled with efficient inference algorithms, provide a very flexible foundation for model-based machine learning, and we outline a large-scale commercial application of this framework involving tens of millions of users. We also describe the concept of probabilistic programming as a powerful software environment for model-based machine learning, and we discuss a specific probabilistic programming language called Infer.NET, which has been widely used in practical applications. PMID:23277612

  11. Model-based machine learning.

    PubMed

    Bishop, Christopher M

    2013-02-13

    Several decades of research in the field of machine learning have resulted in a multitude of different algorithms for solving a broad range of problems. To tackle a new application, a researcher typically tries to map their problem onto one of these existing methods, often influenced by their familiarity with specific algorithms and by the availability of corresponding software implementations. In this study, we describe an alternative methodology for applying machine learning, in which a bespoke solution is formulated for each new application. The solution is expressed through a compact modelling language, and the corresponding custom machine learning code is then generated automatically. This model-based approach offers several major advantages, including the opportunity to create highly tailored models for specific scenarios, as well as rapid prototyping and comparison of a range of alternative models. Furthermore, newcomers to the field of machine learning do not have to learn about the huge range of traditional methods, but instead can focus their attention on understanding a single modelling environment. In this study, we show how probabilistic graphical models, coupled with efficient inference algorithms, provide a very flexible foundation for model-based machine learning, and we outline a large-scale commercial application of this framework involving tens of millions of users. We also describe the concept of probabilistic programming as a powerful software environment for model-based machine learning, and we discuss a specific probabilistic programming language called Infer.NET, which has been widely used in practical applications.

  12. Integrating model abstraction into monitoring strategies

    USDA-ARS?s Scientific Manuscript database

    This study was designed and performed to investigate the opportunities and benefits of integrating model abstraction techniques into monitoring strategies. The study focused on future applications of modeling to contingency planning and management of potential and actual contaminant release sites wi...

  13. Directory of Energy Information Administration Model Abstracts

    SciTech Connect

    Not Available

    1986-07-16

    This directory partially fulfills the requirements of Section 8c, of the documentation order, which states in part that: The Office of Statistical Standards will annually publish an EIA document based on the collected abstracts and the appendices. This report contains brief statements about each model's title, acronym, purpose, and status, followed by more detailed information on characteristics, uses, and requirements. Sources for additional information are identified. All models active through March 1985 are included. The main body of this directory is an alphabetical list of all active EIA models. Appendix A identifies major EIA modeling systems and the models withinmore » these systems, and Appendix B identifies active EIA models by type (basic, auxiliary, and developing). EIA also leases models developed by proprietary software vendors. Documentation for these proprietary models is the responsibility of the companies from which they are leased. EIA has recently leased models from Chase Econometrics, Inc., Data Resources, Inc. (DRI), the Oak Ridge National Laboratory (ORNL), and Wharton Econometric Forecasting Associates (WEFA). Leased models are not abstracted here. The directory is intended for the use of energy and energy-policy analysts in the public and private sectors.« less

  14. Directory of Energy Information Administration model abstracts

    SciTech Connect

    Not Available

    1987-08-11

    This report contains brief statements from the model managers about each model's title, acronym, purpose, and status, followed by more detailed information on characteristics, uses, and requirements. Sources for additional information are identified. All models ''active'' through March 1987 are included. The main body of this directory is an alphabetical list of all active EIA models. Appendix A identifies major EIA modeling systems and the models within these systems, and Appendix B identifies active EIA models by type (basic, auxiliary, and developing). A basic model is one designated by the EIA Administrator as being sufficiently important to require sustained supportmore » and public scrutiny. An auxiliary model is one designated by the EIA Administrator as being used only occasionally in analyses, and therefore requires minimal levels of documentation. A developing model is one designated by the EIA Administrator as being under development and yet of sufficient interest to require a basic level of documentation at a future date. EIA also leases models developed by proprietary software vendors. Documentation for these ''proprietary'' models is the responsibility of the companies from which they are leased. EIA has recently leased models from Chase Econometrics, Inc., Data Resources, Inc. (DRI), the Oak Ridge National Laboratory (ORNL), and Wharton Econometric Forecasting Associates (WEFA). Leased models are not abstracted here. The directory is intended for the use of energy and energy-policy analysts in the public and private sectors.« less

  15. Model Checking Abstract PLEXIL Programs with SMART

    NASA Technical Reports Server (NTRS)

    Siminiceanu, Radu I.

    2007-01-01

    We describe a method to automatically generate discrete-state models of abstract Plan Execution Interchange Language (PLEXIL) programs that can be analyzed using model checking tools. Starting from a high-level description of a PLEXIL program or a family of programs with common characteristics, the generator lays the framework that models the principles of program execution. The concrete parts of the program are not automatically generated, but require the modeler to introduce them by hand. As a case study, we generate models to verify properties of the PLEXIL macro constructs that are introduced as shorthand notation. After an exhaustive analysis, we conclude that the macro definitions obey the intended semantics and behave as expected, but contingently on a few specific requirements on the timing semantics of micro-steps in the concrete executive implementation.

  16. Abstracts

    ERIC Educational Resources Information Center

    American Biology Teacher, 1976

    1976-01-01

    Presents abstracts of 63 papers to be presented at the 1976 Convention of the National Association of Biology Teachers, October 14-17, 1976, Denver, Colorado. Papers cover a wide range of biology and science education topics with the majority concentrating upon the convention's main program, "Ecosystems: 1776-1976-?". (SL)

  17. Multivariate cross-classification: applying machine learning techniques to characterize abstraction in neural representations

    PubMed Central

    Kaplan, Jonas T.; Man, Kingson; Greening, Steven G.

    2015-01-01

    Here we highlight an emerging trend in the use of machine learning classifiers to test for abstraction across patterns of neural activity. When a classifier algorithm is trained on data from one cognitive context, and tested on data from another, conclusions can be drawn about the role of a given brain region in representing information that abstracts across those cognitive contexts. We call this kind of analysis Multivariate Cross-Classification (MVCC), and review several domains where it has recently made an impact. MVCC has been important in establishing correspondences among neural patterns across cognitive domains, including motor-perception matching and cross-sensory matching. It has been used to test for similarity between neural patterns evoked by perception and those generated from memory. Other work has used MVCC to investigate the similarity of representations for semantic categories across different kinds of stimulus presentation, and in the presence of different cognitive demands. We use these examples to demonstrate the power of MVCC as a tool for investigating neural abstraction and discuss some important methodological issues related to its application. PMID:25859202

  18. Model Checking, Abstraction, and Compositional Verification

    DTIC Science & Technology

    1993-07-01

    the ( alois connections used by Bensalrnu el al. [6], and also has some relation to Kurshan’s automata homonuor- phisms [62]. (Actually. we can impose a...multiprocessor simulation model. ACM Transactions on Computer Systems, 4(4):273-298, November 1986. [41 D. L. Beatty, R. E. Bryant, and C.-J. Seger

  19. Application of model abstraction techniques to simulate transport in soils

    USDA-ARS?s Scientific Manuscript database

    Successful understanding and modeling of contaminant transport in soils is the precondition of risk-informed predictions of the subsurface contaminant transport. Exceedingly complex models of subsurface contaminant transport are often inefficient. Model abstraction is the methodology for reducing th...

  20. Abstracting event-based control models for high autonomy systems

    NASA Technical Reports Server (NTRS)

    Luh, Cheng-Jye; Zeigler, Bernard P.

    1993-01-01

    A high autonomy system needs many models on which to base control, management, design, and other interventions. These models differ in level of abstraction and in formalism. Concepts and tools are needed to organize the models into a coherent whole. The paper deals with the abstraction processes for systematic derivation of related models for use in event-based control. The multifaceted modeling methodology is briefly reviewed. The morphism concepts needed for application to model abstraction are described. A theory for supporting the construction of DEVS models needed for event-based control is then presented. An implemented morphism on the basis of this theory is also described.

  1. Relative Effectiveness of Titles, Abstracts, and Subject Headings for Machine Retrieval from the COMPENDEX Services

    ERIC Educational Resources Information Center

    Byrne, Jerry R.

    1975-01-01

    Investigated the relative merits of searching on titles, subject headings, abstracts, free-language terms, and combinations of these elements. The combination of titles and abstracts came the closest to 100 percent retrieval. (Author/PF)

  2. Modelling abstraction licensing strategies ahead of the UK's water abstraction licensing reform

    NASA Astrophysics Data System (ADS)

    Klaar, M. J.

    2012-12-01

    Within England and Wales, river water abstractions are licensed and regulated by the Environment Agency (EA), who uses compliance with the Environmental Flow Indicator (EFI) to ascertain where abstraction may cause undesirable effects on river habitats and species. The EFI is a percentage deviation from natural flow represented using a flow duration curve. The allowable percentage deviation changes with different flows, and also changes depending on an assessment of the sensitivity of the river to changes in flow (Table 1). Within UK abstraction licensing, resource availability is expressed as a surplus or deficit of water resources in relation to the EFI, and utilises the concept of 'hands-off-flows' (HOFs) at the specified flow statistics detailed in Table 1. Use of a HOF system enables abstraction to cease at set flows, but also enables abstraction to occur at periods of time when more water is available. Compliance at low flows (Q95) is used by the EA to determine the hydrological classification and compliance with the Water Framework Directive (WFD) for identifying waterbodies where flow may be causing or contributing to a failure in good ecological status (GES; Table 2). This compliance assessment shows where the scenario flows are below the EFI and by how much, to help target measures for further investigation and assessment. Currently, the EA is reviewing the EFI methodology in order to assess whether or not it can be used within the reformed water abstraction licensing system which is being planned by the Department for Environment, Food and Rural Affairs (DEFRA) to ensure the licensing system is resilient to the challenges of climate change and population growth, while allowing abstractors to meet their water needs efficiently, and better protect the environment. In order to assess the robustness of the EFI, a simple model has been created which allows a number of abstraction, flow and licensing scenarios to be run to determine WFD compliance using the

  3. Coupling Radar Rainfall to Hydrological Models for Water Abstraction Management

    NASA Astrophysics Data System (ADS)

    Asfaw, Alemayehu; Shucksmith, James; Smith, Andrea; MacDonald, Ken

    2015-04-01

    The impacts of climate change and growing water use are likely to put considerable pressure on water resources and the environment. In the UK, a reform to surface water abstraction policy has recently been proposed which aims to increase the efficiency of using available water resources whilst minimising impacts on the aquatic environment. Key aspects to this reform include the consideration of dynamic rather than static abstraction licensing as well as introducing water trading concepts. Dynamic licensing will permit varying levels of abstraction dependent on environmental conditions (i.e. river flow and quality). The practical implementation of an effective dynamic abstraction strategy requires suitable flow forecasting techniques to inform abstraction asset management. Potentially the predicted availability of water resources within a catchment can be coupled to predicted demand and current storage to inform a cost effective water resource management strategy which minimises environmental impacts. The aim of this work is to use a historical analysis of UK case study catchment to compare potential water resource availability using modelled dynamic abstraction scenario informed by a flow forecasting model, against observed abstraction under a conventional abstraction regime. The work also demonstrates the impacts of modelling uncertainties on the accuracy of predicted water availability over range of forecast lead times. The study utilised a conceptual rainfall-runoff model PDM - Probability-Distributed Model developed by Centre for Ecology & Hydrology - set up in the Dove River catchment (UK) using 1km2 resolution radar rainfall as inputs and 15 min resolution gauged flow data for calibration and validation. Data assimilation procedures are implemented to improve flow predictions using observed flow data. Uncertainties in the radar rainfall data used in the model are quantified using artificial statistical error model described by Gaussian distribution and

  4. How Pupils Use a Model for Abstract Concepts in Genetics

    ERIC Educational Resources Information Center

    Venville, Grady; Donovan, Jenny

    2008-01-01

    The purpose of this research was to explore the way pupils of different age groups use a model to understand abstract concepts in genetics. Pupils from early childhood to late adolescence were taught about genes and DNA using an analogical model (the wool model) during their regular biology classes. Changing conceptual understandings of the…

  5. Concrete Model Checking with Abstract Matching and Refinement

    NASA Technical Reports Server (NTRS)

    Pasareanu Corina S.; Peianek Radek; Visser, Willem

    2005-01-01

    We propose an abstraction-based model checking method which relies on refinement of an under-approximation of the feasible behaviors of the system under analysis. The method preserves errors to safety properties, since all analyzed behaviors are feasible by definition. The method does not require an abstract transition relation to he generated, but instead executes the concrete transitions while storing abstract versions of the concrete states, as specified by a set of abstraction predicates. For each explored transition. the method checks, with the help of a theorem prover, whether there is any loss of precision introduced by abstraction. The results of these checks are used to decide termination or to refine the abstraction, by generating new abstraction predicates. If the (possibly infinite) concrete system under analysis has a finite bisimulation quotient, then the method is guaranteed to eventually explore an equivalent finite bisimilar structure. We illustrate the application of the approach for checking concurrent programs. We also show how a lightweight variant can be used for efficient software testing.

  6. Do Concretely and Abstractly Worded Arguments Require Different Models?

    ERIC Educational Resources Information Center

    Hample, Dale

    Dale Hample's cognitive model of argument is designed to reflect the operation of syllogistic thought processes. It has been suggested however, that the model applies more closely to abstractly worded arguments than to concrete thinking and that it also may work better with more interested respondents because it seems to describe the central…

  7. An abstract specification language for Markov reliability models

    NASA Technical Reports Server (NTRS)

    Butler, R. W.

    1985-01-01

    Markov models can be used to compute the reliability of virtually any fault tolerant system. However, the process of delineating all of the states and transitions in a model of complex system can be devastatingly tedious and error-prone. An approach to this problem is presented utilizing an abstract model definition language. This high level language is described in a nonformal manner and illustrated by example.

  8. An abstract language for specifying Markov reliability models

    NASA Technical Reports Server (NTRS)

    Butler, Ricky W.

    1986-01-01

    Markov models can be used to compute the reliability of virtually any fault tolerant system. However, the process of delineating all of the states and transitions in a model of complex system can be devastatingly tedious and error-prone. An approach to this problem is presented utilizing an abstract model definition language. This high level language is described in a nonformal manner and illustrated by example.

  9. Sensing position and speed by recording magnetization transitions on mechanically functional machine members (abstract)

    NASA Astrophysics Data System (ADS)

    Garshelis, I. J.

    1997-04-01

    Conventional means of sensing position and speed of moving machine members for control purposes typically requires the use of supplementary, ad hoc devices. Many mechanically functional moving machine members are fabricated from ferromagnetic steels and may, thus, provide an opportunity to themselves carry positionally relevant information in the form of local regions of deliberately instilled remanent magnetization, Mr. To avoid ambiguities associated with the imprecise borders of such regions as well as their possibly preexisting presence, information is more reliably carried in the form of local transitions in the polarity of Mr from a quiescent bias. The presence and physical location of such transitions relative to reference features either on the member itself or on other members undergoing correlated motion constitutes stored information. The presence of a transition is signaled by the transitory appearance of the external field associated with ∇ṡMr as the transition containing region passes by a magnetic-field detecting device fixed to the machine frame. Implanting and removing transitions from parts while in motion is readily accomplished by pulsed currents and biasing magnets. While the whole process of storing, reading, and erasing bits of information in magnetic form follows the concepts and principles of conventional magnetic recording, profoundly different quantitative factors, conditions, and performance requirements affect the implementation of the described sensing system. In particular, the coercivity, Hc, of commonly used steels is 3-30 Oe versus 300-1200 Oe in recording media and both the thickness of the media and the air gaps separating the media surface from the heads used in conventional systems are each 2-3 orders of magnitude smaller than their counterparts in the described system, where speed may also be variable down to zero. While the combined effect of these factors is to greatly diminish the attainable density of recorded

  10. An abstract approach to evaporation models in rarefied gas dynamics

    NASA Astrophysics Data System (ADS)

    Greenberg, W.; van der Mee, C. V. M.

    1984-03-01

    Strong evaporation models involving 1D stationary problems with linear self-adjoint collision operators and solutions in abstract Hilbert spaces are investigated analytically. An efficient algorithm for locating the transition from existence to nonexistence of solutions is developed and applied to the 1D and 3D BGK model equations and the 3D BGK model in moment form, demonstrating the nonexistence of stationary evaporation states with supersonic drift velocities. Applications to similar models in electron and phonon transport, radiative transfer, and neutron transport are suggested.

  11. Modeling Electronic Quantum Transport with Machine Learning

    DOE PAGES

    Lopez Bezanilla, Alejandro; von Lilienfeld Toal, Otto A.

    2014-06-11

    We present a machine learning approach to solve electronic quantum transport equations of one-dimensional nanostructures. The transmission coefficients of disordered systems were computed to provide training and test data sets to the machine. The system’s representation encodes energetic as well as geometrical information to characterize similarities between disordered configurations, while the Euclidean norm is used as a measure of similarity. Errors for out-of-sample predictions systematically decrease with training set size, enabling the accurate and fast prediction of new transmission coefficients. The remarkable performance of our model to capture the complexity of interference phenomena lends further support to its viability inmore » dealing with transport problems of undulatory nature.« less

  12. Towards a generalized energy prediction model for machine tools

    PubMed Central

    Bhinge, Raunak; Park, Jinkyoo; Law, Kincho H.; Dornfeld, David A.; Helu, Moneer; Rachuri, Sudarsan

    2017-01-01

    Energy prediction of machine tools can deliver many advantages to a manufacturing enterprise, ranging from energy-efficient process planning to machine tool monitoring. Physics-based, energy prediction models have been proposed in the past to understand the energy usage pattern of a machine tool. However, uncertainties in both the machine and the operating environment make it difficult to predict the energy consumption of the target machine reliably. Taking advantage of the opportunity to collect extensive, contextual, energy-consumption data, we discuss a data-driven approach to develop an energy prediction model of a machine tool in this paper. First, we present a methodology that can efficiently and effectively collect and process data extracted from a machine tool and its sensors. We then present a data-driven model that can be used to predict the energy consumption of the machine tool for machining a generic part. Specifically, we use Gaussian Process (GP) Regression, a non-parametric machine-learning technique, to develop the prediction model. The energy prediction model is then generalized over multiple process parameters and operations. Finally, we apply this generalized model with a method to assess uncertainty intervals to predict the energy consumed to machine any part using a Mori Seiki NVD1500 machine tool. Furthermore, the same model can be used during process planning to optimize the energy-efficiency of a machining process. PMID:28652687

  13. Towards a generalized energy prediction model for machine tools.

    PubMed

    Bhinge, Raunak; Park, Jinkyoo; Law, Kincho H; Dornfeld, David A; Helu, Moneer; Rachuri, Sudarsan

    2017-04-01

    Energy prediction of machine tools can deliver many advantages to a manufacturing enterprise, ranging from energy-efficient process planning to machine tool monitoring. Physics-based, energy prediction models have been proposed in the past to understand the energy usage pattern of a machine tool. However, uncertainties in both the machine and the operating environment make it difficult to predict the energy consumption of the target machine reliably. Taking advantage of the opportunity to collect extensive, contextual, energy-consumption data, we discuss a data-driven approach to develop an energy prediction model of a machine tool in this paper. First, we present a methodology that can efficiently and effectively collect and process data extracted from a machine tool and its sensors. We then present a data-driven model that can be used to predict the energy consumption of the machine tool for machining a generic part. Specifically, we use Gaussian Process (GP) Regression, a non-parametric machine-learning technique, to develop the prediction model. The energy prediction model is then generalized over multiple process parameters and operations. Finally, we apply this generalized model with a method to assess uncertainty intervals to predict the energy consumed to machine any part using a Mori Seiki NVD1500 machine tool. Furthermore, the same model can be used during process planning to optimize the energy-efficiency of a machining process.

  14. A Machine Learning Approach to Student Modeling.

    DTIC Science & Technology

    1984-05-01

    machine learning , and describe ACN, a student modeling system that incorporates this approach. This system begins with a set of overly general rules, which it uses to search a problem space until it arrives at the same answer as the student. The ACM computer program then uses the solution path it has discovered to determine positive and negative instances of its initial rules, and employs a discrimination learning mechanism to place additional conditions on these rules. The revised rules will reproduce the solution path without search, and constitute a cognitive model of

  15. Derivation of Rigid Body Analysis Models from Vehicle Architecture Abstractions

    DTIC Science & Technology

    2011-06-17

    models of every type have their basis in some type of physical representation of the design domain. Rather than describing three-dimensional continua of...arrangement, while capturing just enough physical detail to be used as the basis for a meaningful representation of the design , and eventually, analyses that...permit architecture assessment. The design information captured by the abstractions is available at the very earliest stages of the vehicle

  16. Cerebellar input configuration toward object model abstraction in manipulation tasks.

    PubMed

    Luque, Niceto R; Garrido, Jesus A; Carrillo, Richard R; Coenen, Olivier J-M D; Ros, Eduardo

    2011-08-01

    It is widely assumed that the cerebellum is one of the main nervous centers involved in correcting and refining planned movement and accounting for disturbances occurring during movement, for instance, due to the manipulation of objects which affect the kinematics and dynamics of the robot-arm plant model. In this brief, we evaluate a way in which a cerebellar-like structure can store a model in the granular and molecular layers. Furthermore, we study how its microstructure and input representations (context labels and sensorimotor signals) can efficiently support model abstraction toward delivering accurate corrective torque values for increasing precision during different-object manipulation. We also describe how the explicit (object-related input labels) and implicit state input representations (sensorimotor signals) complement each other to better handle different models and allow interpolation between two already stored models. This facilitates accurate corrections during manipulations of new objects taking advantage of already stored models.

  17. Directory of Energy Information Administration model abstracts 1988

    SciTech Connect

    Not Available

    1988-01-01

    This directory contains descriptions about each basic and auxiliary model, including the title, acronym, purpose, and type, followed by more detailed information on characteristics, uses, and requirements. For developing models, limited information is provided. Sources for additional information are identified. Included in this directory are 44 EIA models active as of February 1, 1988; 16 of which operate on personal computers. Models that run on personal computers are identified by ''PC'' as part of the acronyms. The main body of this directory is an alphabetical listing of all basic and auxiliary EIA models. Appendix A identifies major EIA modeling systemsmore » and the models within these systems, and Appendix B identifies EIA models by type (basic or auxiliary). Appendix C lists developing models and contact persons for those models. A basic model is one designated by the EIA Administrator as being sufficiently important to require sustained support and public scrutiny. An auxiliary model is one designated by the EIA Administrator as being used only occasionally in analyses, and therefore requires minimal levels of documentation. A developing model is one designated by the EIA Administrator as being under development and yet of sufficient interest to require a basic level of documentation at a future date. EIA also leases models developed by proprietary software vendors. Documentation for these ''proprietary'' models is the responsibility of the companies from which they are leased. EIA has recently leased models from Chase Econometrics, Inc., Data Resources, Inc. (DRI), the Oak Ridge National Laboratory (ORNL), and Wharton Econometric Forecasting Associates (WEFA). Leased models are not abstracted here.« less

  18. Entity-Centric Abstraction and Modeling Framework for Transportation Architectures

    NASA Technical Reports Server (NTRS)

    Lewe, Jung-Ho; DeLaurentis, Daniel A.; Mavris, Dimitri N.; Schrage, Daniel P.

    2007-01-01

    A comprehensive framework for representing transpportation architectures is presented. After discussing a series of preceding perspectives and formulations, the intellectual underpinning of the novel framework using an entity-centric abstraction of transportation is described. The entities include endogenous and exogenous factors and functional expressions are offered that relate these and their evolution. The end result is a Transportation Architecture Field which permits analysis of future concepts under the holistic perspective. A simulation model which stems from the framework is presented and exercised producing results which quantify improvements in air transportation due to advanced aircraft technologies. Finally, a modeling hypothesis and its accompanying criteria are proposed to test further use of the framework for evaluating new transportation solutions.

  19. A Machine-Learning-Driven Sky Model.

    PubMed

    Satylmys, Pynar; Bashford-Rogers, Thomas; Chalmers, Alan; Debattista, Kurt

    2017-01-01

    Sky illumination is responsible for much of the lighting in a virtual environment. A machine-learning-based approach can compactly represent sky illumination from both existing analytic sky models and from captured environment maps. The proposed approach can approximate the captured lighting at a significantly reduced memory cost and enable smooth transitions of sky lighting to be created from a small set of environment maps captured at discrete times of day. The author's results demonstrate accuracy close to the ground truth for both analytical and capture-based methods. The approach has a low runtime overhead, so it can be used as a generic approach for both offline and real-time applications.

  20. Workshop on Fielded Applications of Machine Learning Held in Amherst, Massachusetts on 30 June-1 July 1993. Abstracts.

    DTIC Science & Technology

    1993-01-01

    engineering has led to many AI systems that are now regularly used in industry and elsewhere. The ultimate test of machine learning , the subfield of Al that...applications of machine learning suggest the time was ripe for a meeting on this topic. For this reason, Pat Langley (Siemens Corporate Research) and Yves...Kodratoff (Universite de Paris, Sud) organized an invited workshop on applications of machine learning . The goal of the gathering was to familiarize

  1. 9. VIEW, LOOKING SOUTH, OF INTERLOCKING MACHINE, WITH ORIGINAL MODEL ...

    Library of Congress Historic Buildings Survey, Historic Engineering Record, Historic Landscapes Survey

    9. VIEW, LOOKING SOUTH, OF INTERLOCKING MACHINE, WITH ORIGINAL MODEL BOARD IN CENTER, NEW MODEL BOARD AT LEFT AND MODEL SEMAPHORES AT TOP OF PHOTOGRAPH, THIRD FLOOR - South Station Tower No. 1 & Interlocking System, Dewey Square, Boston, Suffolk County, MA

  2. A rule-based approach to model checking of UML state machines

    NASA Astrophysics Data System (ADS)

    Grobelna, Iwona; Grobelny, Michał; Stefanowicz, Łukasz

    2016-12-01

    In the paper a new approach to formal verification of control process specification expressed by means of UML state machines in version 2.x is proposed. In contrast to other approaches from the literature, we use the abstract and universal rule-based logical model suitable both for model checking (using the nuXmv model checker), but also for logical synthesis in form of rapid prototyping. Hence, a prototype implementation in hardware description language VHDL can be obtained that fully reflects the primary, already formally verified specification in form of UML state machines. Presented approach allows to increase the assurance that implemented system meets the user-defined requirements.

  3. Modeling Patient Treatment With Medical Records: An Abstraction Hierarchy to Understand User Competencies and Needs

    PubMed Central

    2017-01-01

    Background Health care is a complex sociotechnical system. Patient treatment is evolving and needs to incorporate the use of technology and new patient-centered treatment paradigms. Cognitive work analysis (CWA) is an effective framework for understanding complex systems, and work domain analysis (WDA) is useful for understanding complex ecologies. Although previous applications of CWA have described patient treatment, due to their scope of work patients were previously characterized as biomedical machines, rather than patient actors involved in their own care. Objective An abstraction hierarchy that characterizes patients as beings with complex social values and priorities is needed. This can help better understand treatment in a modern approach to care. The purpose of this study was to perform a WDA to represent the treatment of patients with medical records. Methods The methods to develop this model included the analysis of written texts and collaboration with subject matter experts. Our WDA represents the ecology through its functional purposes, abstract functions, generalized functions, physical functions, and physical forms. Results Compared with other work domain models, this model is able to articulate the nuanced balance between medical treatment, patient education, and limited health care resources. Concepts in the analysis were similar to the modeling choices of other WDAs but combined them in as a comprehensive, systematic, and contextual overview. The model is helpful to understand user competencies and needs. Future models could be developed to model the patient’s domain and enable the exploration of the shared decision-making (SDM) paradigm. Conclusion Our work domain model links treatment goals, decision-making constraints, and task workflows. This model can be used by system developers who would like to use ecological interface design (EID) to improve systems. Our hierarchy is the first in a future set that could explore new treatment paradigms

  4. Investigation of approximate models of experimental temperature characteristics of machines

    NASA Astrophysics Data System (ADS)

    Parfenov, I. V.; Polyakov, A. N.

    2018-05-01

    This work is devoted to the investigation of various approaches to the approximation of experimental data and the creation of simulation mathematical models of thermal processes in machines with the aim of finding ways to reduce the time of their field tests and reducing the temperature error of the treatments. The main methods of research which the authors used in this work are: the full-scale thermal testing of machines; realization of various approaches at approximation of experimental temperature characteristics of machine tools by polynomial models; analysis and evaluation of modelling results (model quality) of the temperature characteristics of machines and their derivatives up to the third order in time. As a result of the performed researches, rational methods, type, parameters and complexity of simulation mathematical models of thermal processes in machine tools are proposed.

  5. 8. VIEW, LOOKING NORTH, OF INTERLOCKING MACHINE WITH ORIGINAL MODEL ...

    Library of Congress Historic Buildings Survey, Historic Engineering Record, Historic Landscapes Survey

    8. VIEW, LOOKING NORTH, OF INTERLOCKING MACHINE WITH ORIGINAL MODEL BOARD IN CENTER AND MODEL SEMAPHORE SIGNALS (AT TOP OF PHOTOGRAPH), THIRD FLOOR - South Station Tower No. 1 & Interlocking System, Dewey Square, Boston, Suffolk County, MA

  6. Generative Modeling for Machine Learning on the D-Wave

    SciTech Connect

    Thulasidasan, Sunil

    These are slides on Generative Modeling for Machine Learning on the D-Wave. The following topics are detailed: generative models; Boltzmann machines: a generative model; restricted Boltzmann machines; learning parameters: RBM training; practical ways to train RBM; D-Wave as a Boltzmann sampler; mapping RBM onto the D-Wave; Chimera restricted RBM; mapping binary RBM to Ising model; experiments; data; D-Wave effective temperature, parameters noise, etc.; experiments: contrastive divergence (CD) 1 step; after 50 steps of CD; after 100 steps of CD; D-Wave (experiments 1, 2, 3); D-Wave observations.

  7. Editors' Introduction: Abstract Concepts: Structure, Processing, and Modeling.

    PubMed

    Bolognesi, Marianna; Steen, Gerard

    2018-06-22

    Our ability to deal with abstract concepts is one of the most intriguing faculties of human cognition. Still, we know little about how such concepts are formed, processed, and represented in mind. For example, because abstract concepts do not designate referents that can be experienced through our body, the role of perceptual experiences in shaping their content remains controversial. Current theories suggest a variety of alternative explanations to the question of "how abstract concepts are represented in the human mind." These views pinpoint specific streams of semantic information that would play a prominent role in shaping the content of abstract concepts, such as situation-based information (e.g., Barsalou & Wiemer-Hastings, ), affective information (Kousta, Vigliocco, Vinson, Andrews, & Del Campo, ), and linguistic information (Louwerse, ). Rarely, these theoretical views are directly compared. In this special issue, current views are presented in their most recent and advanced form, and directly compared and discussed in a debate, which is reported at the end of each article. As a result, new exciting questions and challenges arise. These questions and challenges, reported in this introductory article, can arguably pave the way to new empirical studies and theoretical developments on the nature of abstract concepts. © 2018 Cognitive Science Society, Inc.

  8. Model Machine Shop for Drafting Instruction.

    ERIC Educational Resources Information Center

    Jackson, Carl R.

    The development and implementation of a two-year interdisciplinary course integrating a machine shop and drafting curriculum are described in the report. The purpose of the course is to provide a learning process in industrial drafting featuring identifiable orientation in skills that will enable the student to develop competencies that are…

  9. Developing Parametric Models for the Assembly of Machine Fixtures for Virtual Multiaxial CNC Machining Centers

    NASA Astrophysics Data System (ADS)

    Balaykin, A. V.; Bezsonov, K. A.; Nekhoroshev, M. V.; Shulepov, A. P.

    2018-01-01

    This paper dwells upon a variance parameterization method. Variance or dimensional parameterization is based on sketching, with various parametric links superimposed on the sketch objects and user-imposed constraints in the form of an equation system that determines the parametric dependencies. This method is fully integrated in a top-down design methodology to enable the creation of multi-variant and flexible fixture assembly models, as all the modeling operations are hierarchically linked in the built tree. In this research the authors consider a parameterization method of machine tooling used for manufacturing parts using multiaxial CNC machining centers in the real manufacturing process. The developed method allows to significantly reduce tooling design time when making changes of a part’s geometric parameters. The method can also reduce time for designing and engineering preproduction, in particular, for development of control programs for CNC equipment and control and measuring machines, automate the release of design and engineering documentation. Variance parameterization helps to optimize construction of parts as well as machine tooling using integrated CAE systems. In the framework of this study, the authors demonstrate a comprehensive approach to parametric modeling of machine tooling in the CAD package used in the real manufacturing process of aircraft engines.

  10. On the Conditioning of Machine-Learning-Assisted Turbulence Modeling

    NASA Astrophysics Data System (ADS)

    Wu, Jinlong; Sun, Rui; Wang, Qiqi; Xiao, Heng

    2017-11-01

    Recently, several researchers have demonstrated that machine learning techniques can be used to improve the RANS modeled Reynolds stress by training on available database of high fidelity simulations. However, obtaining improved mean velocity field remains an unsolved challenge, restricting the predictive capability of current machine-learning-assisted turbulence modeling approaches. In this work we define a condition number to evaluate the model conditioning of data-driven turbulence modeling approaches, and propose a stability-oriented machine learning framework to model Reynolds stress. Two canonical flows, the flow in a square duct and the flow over periodic hills, are investigated to demonstrate the predictive capability of the proposed framework. The satisfactory prediction performance of mean velocity field for both flows demonstrates the predictive capability of the proposed framework for machine-learning-assisted turbulence modeling. With showing the capability of improving the prediction of mean flow field, the proposed stability-oriented machine learning framework bridges the gap between the existing machine-learning-assisted turbulence modeling approaches and the demand of predictive capability of turbulence models in real applications.

  11. An Abstraction-Based Data Model for Information Retrieval

    NASA Astrophysics Data System (ADS)

    McAllister, Richard A.; Angryk, Rafal A.

    Language ontologies provide an avenue for automated lexical analysis that may be used to supplement existing information retrieval methods. This paper presents a method of information retrieval that takes advantage of WordNet, a lexical database, to generate paths of abstraction, and uses them as the basis for an inverted index structure to be used in the retrieval of documents from an indexed corpus. We present this method as a entree to a line of research on using ontologies to perform word-sense disambiguation and improve the precision of existing information retrieval techniques.

  12. Context in Models of Human-Machine Systems

    NASA Technical Reports Server (NTRS)

    Callantine, Todd J.; Null, Cynthia H. (Technical Monitor)

    1998-01-01

    All human-machine systems models represent context. This paper proposes a theory of context through which models may be usefully related and integrated for design. The paper presents examples of context representation in various models, describes an application to developing models for the Crew Activity Tracking System (CATS), and advances context as a foundation for integrated design of complex dynamic systems.

  13. Predicting Market Impact Costs Using Nonparametric Machine Learning Models.

    PubMed

    Park, Saerom; Lee, Jaewook; Son, Youngdoo

    2016-01-01

    Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance.

  14. Predicting Market Impact Costs Using Nonparametric Machine Learning Models

    PubMed Central

    Park, Saerom; Lee, Jaewook; Son, Youngdoo

    2016-01-01

    Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance. PMID:26926235

  15. Network Modeling and Energy-Efficiency Optimization for Advanced Machine-to-Machine Sensor Networks

    PubMed Central

    Jung, Sungmo; Kim, Jong Hyun; Kim, Seoksoo

    2012-01-01

    Wireless machine-to-machine sensor networks with multiple radio interfaces are expected to have several advantages, including high spatial scalability, low event detection latency, and low energy consumption. Here, we propose a network model design method involving network approximation and an optimized multi-tiered clustering algorithm that maximizes node lifespan by minimizing energy consumption in a non-uniformly distributed network. Simulation results show that the cluster scales and network parameters determined with the proposed method facilitate a more efficient performance compared to existing methods. PMID:23202190

  16. Developing a PLC-friendly state machine model: lessons learned

    NASA Astrophysics Data System (ADS)

    Pessemier, Wim; Deconinck, Geert; Raskin, Gert; Saey, Philippe; Van Winckel, Hans

    2014-07-01

    Modern Programmable Logic Controllers (PLCs) have become an attractive platform for controlling real-time aspects of astronomical telescopes and instruments due to their increased versatility, performance and standardization. Likewise, vendor-neutral middleware technologies such as OPC Unified Architecture (OPC UA) have recently demonstrated that they can greatly facilitate the integration of these industrial platforms into the overall control system. Many practical questions arise, however, when building multi-tiered control systems that consist of PLCs for low level control, and conventional software and platforms for higher level control. How should the PLC software be structured, so that it can rely on well-known programming paradigms on the one hand, and be mapped to a well-organized OPC UA interface on the other hand? Which programming languages of the IEC 61131-3 standard closely match the problem domains of the abstraction levels within this structure? How can the recent additions to the standard (such as the support for namespaces and object-oriented extensions) facilitate a model based development approach? To what degree can our applications already take advantage of the more advanced parts of the OPC UA standard, such as the high expressiveness of the semantic modeling language that it defines, or the support for events, aggregation of data, automatic discovery, ... ? What are the timing and concurrency problems to be expected for the higher level tiers of the control system due to the cyclic execution of control and communication tasks by the PLCs? We try to answer these questions by demonstrating a semantic state machine model that can readily be implemented using IEC 61131 and OPC UA. One that does not aim to capture all possible states of a system, but rather one that attempts to organize the course-grained structure and behaviour of a system. In this paper we focus on the intricacies of this seemingly simple task, and on the lessons that we

  17. Modeling Patient Treatment With Medical Records: An Abstraction Hierarchy to Understand User Competencies and Needs.

    PubMed

    St-Maurice, Justin D; Burns, Catherine M

    2017-07-28

    Health care is a complex sociotechnical system. Patient treatment is evolving and needs to incorporate the use of technology and new patient-centered treatment paradigms. Cognitive work analysis (CWA) is an effective framework for understanding complex systems, and work domain analysis (WDA) is useful for understanding complex ecologies. Although previous applications of CWA have described patient treatment, due to their scope of work patients were previously characterized as biomedical machines, rather than patient actors involved in their own care. An abstraction hierarchy that characterizes patients as beings with complex social values and priorities is needed. This can help better understand treatment in a modern approach to care. The purpose of this study was to perform a WDA to represent the treatment of patients with medical records. The methods to develop this model included the analysis of written texts and collaboration with subject matter experts. Our WDA represents the ecology through its functional purposes, abstract functions, generalized functions, physical functions, and physical forms. Compared with other work domain models, this model is able to articulate the nuanced balance between medical treatment, patient education, and limited health care resources. Concepts in the analysis were similar to the modeling choices of other WDAs but combined them in as a comprehensive, systematic, and contextual overview. The model is helpful to understand user competencies and needs. Future models could be developed to model the patient's domain and enable the exploration of the shared decision-making (SDM) paradigm. Our work domain model links treatment goals, decision-making constraints, and task workflows. This model can be used by system developers who would like to use ecological interface design (EID) to improve systems. Our hierarchy is the first in a future set that could explore new treatment paradigms. Future hierarchies could model the patient as a

  18. (abstract) Simple Spreadsheet Thermal Models for Cryogenic Applications

    NASA Technical Reports Server (NTRS)

    Nash, A. E.

    1994-01-01

    Self consistent circuit analog thermal models, that can be run in commercial spreadsheet programs on personal computers, have been created to calculate the cooldown and steady state performance of cryogen cooled Dewars. The models include temperature dependent conduction and radiation effects. The outputs of the models provide temperature distribution and Dewar performance information. These models have been used to analyze the Cryogenic Telescope Test Facility (CTTF). The facility will be on line in early 1995 for its first user, the Infrared Telescope Technology Testbed (ITTT), for the Space Infrared Telescope Facility (SIRTF) at JPL. The model algorithm as well as a comparison of the model predictions and actual performance of this facility will be presented.

  19. Symbolic LTL Compilation for Model Checking: Extended Abstract

    NASA Technical Reports Server (NTRS)

    Rozier, Kristin Y.; Vardi, Moshe Y.

    2007-01-01

    In Linear Temporal Logic (LTL) model checking, we check LTL formulas representing desired behaviors against a formal model of the system designed to exhibit these behaviors. To accomplish this task, the LTL formulas must be translated into automata [21]. We focus on LTL compilation by investigating LTL satisfiability checking via a reduction to model checking. Having shown that symbolic LTL compilation algorithms are superior to explicit automata construction algorithms for this task [16], we concentrate here on seeking a better symbolic algorithm.We present experimental data comparing algorithmic variations such as normal forms, encoding methods, and variable ordering and examine their effects on performance metrics including processing time and scalability. Safety critical systems, such as air traffic control, life support systems, hazardous environment controls, and automotive control systems, pervade our daily lives, yet testing and simulation alone cannot adequately verify their reliability [3]. Model checking is a promising approach to formal verification for safety critical systems which involves creating a formal mathematical model of the system and translating desired safety properties into a formal specification for this model. The complement of the specification is then checked against the system model. When the model does not satisfy the specification, model-checking tools accompany this negative answer with a counterexample, which points to an inconsistency between the system and the desired behaviors and aids debugging efforts.

  20. Modelling machine ensembles with discrete event dynamical system theory

    NASA Technical Reports Server (NTRS)

    Hunter, Dan

    1990-01-01

    Discrete Event Dynamical System (DEDS) theory can be utilized as a control strategy for future complex machine ensembles that will be required for in-space construction. The control strategy involves orchestrating a set of interactive submachines to perform a set of tasks for a given set of constraints such as minimum time, minimum energy, or maximum machine utilization. Machine ensembles can be hierarchically modeled as a global model that combines the operations of the individual submachines. These submachines are represented in the global model as local models. Local models, from the perspective of DEDS theory , are described by the following: a set of system and transition states, an event alphabet that portrays actions that takes a submachine from one state to another, an initial system state, a partial function that maps the current state and event alphabet to the next state, and the time required for the event to occur. Each submachine in the machine ensemble is presented by a unique local model. The global model combines the local models such that the local models can operate in parallel under the additional logistic and physical constraints due to submachine interactions. The global model is constructed from the states, events, event functions, and timing requirements of the local models. Supervisory control can be implemented in the global model by various methods such as task scheduling (open-loop control) or implementing a feedback DEDS controller (closed-loop control).

  1. Functional Abstraction from Structure in VLSI Simulation Models,

    DTIC Science & Technology

    1987-05-01

    wide vari- ety of powerful tools, designed around the Y model proposed by Gajski and Kuhn [11]. The heart of the system is the data representation...34Fuictional Models for VLSI Design", 20th IEEE Design Automation Conference (DAC󈨗), 1983, paper 32.2, pp. 506-514. * 21 [11] Gajski , Daniel D., Kuhn, Robert H

  2. Learning About Climate and Atmospheric Models Through Machine Learning

    NASA Astrophysics Data System (ADS)

    Lucas, D. D.

    2017-12-01

    From the analysis of ensemble variability to improving simulation performance, machine learning algorithms can play a powerful role in understanding the behavior of atmospheric and climate models. To learn about model behavior, we create training and testing data sets through ensemble techniques that sample different model configurations and values of input parameters, and then use supervised machine learning to map the relationships between the inputs and outputs. Following this procedure, we have used support vector machines, random forests, gradient boosting and other methods to investigate a variety of atmospheric and climate model phenomena. We have used machine learning to predict simulation crashes, estimate the probability density function of climate sensitivity, optimize simulations of the Madden Julian oscillation, assess the impacts of weather and emissions uncertainty on atmospheric dispersion, and quantify the effects of model resolution changes on precipitation. This presentation highlights recent examples of our applications of machine learning to improve the understanding of climate and atmospheric models. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

  3. Applying model abstraction techniques to optimize monitoring networks for detecting subsurface contaminant transport

    USDA-ARS?s Scientific Manuscript database

    Improving strategies for monitoring subsurface contaminant transport includes performance comparison of competing models, developed independently or obtained via model abstraction. Model comparison and parameter discrimination involve specific performance indicators selected to better understand s...

  4. Phase Transitions in a Model of Y-Molecules Abstract

    NASA Astrophysics Data System (ADS)

    Holz, Danielle; Ruth, Donovan; Toral, Raul; Gunton, James

    Immunoglobulin is a Y-shaped molecule that functions as an antibody to neutralize pathogens. In special cases where there is a high concentration of immunoglobulin molecules, self-aggregation can occur and the molecules undergo phase transitions. This prevents the molecules from completing their function. We used a simplified model of 2-Dimensional Y-molecules with three identical arms on a triangular lattice with 2-dimensional Grand Canonical Ensemble. The molecules were permitted to be placed, removed, rotated or moved on the lattice. Once phase coexistence was found, we used histogram reweighting and multicanonical sampling to calculate our phase diagram.

  5. Modeling Music Emotion Judgments Using Machine Learning Methods.

    PubMed

    Vempala, Naresh N; Russo, Frank A

    2017-01-01

    Emotion judgments and five channels of physiological data were obtained from 60 participants listening to 60 music excerpts. Various machine learning (ML) methods were used to model the emotion judgments inclusive of neural networks, linear regression, and random forests. Input for models of perceived emotion consisted of audio features extracted from the music recordings. Input for models of felt emotion consisted of physiological features extracted from the physiological recordings. Models were trained and interpreted with consideration of the classic debate in music emotion between cognitivists and emotivists. Our models supported a hybrid position wherein emotion judgments were influenced by a combination of perceived and felt emotions. In comparing the different ML approaches that were used for modeling, we conclude that neural networks were optimal, yielding models that were flexible as well as interpretable. Inspection of a committee machine, encompassing an ensemble of networks, revealed that arousal judgments were predominantly influenced by felt emotion, whereas valence judgments were predominantly influenced by perceived emotion.

  6. Technology-assisted title and abstract screening for systematic reviews: a retrospective evaluation of the Abstrackr machine learning tool.

    PubMed

    Gates, Allison; Johnson, Cydney; Hartling, Lisa

    2018-03-12

    Machine learning tools can expedite systematic review (SR) processes by semi-automating citation screening. Abstrackr semi-automates citation screening by predicting relevant records. We evaluated its performance for four screening projects. We used a convenience sample of screening projects completed at the Alberta Research Centre for Health Evidence, Edmonton, Canada: three SRs and one descriptive analysis for which we had used SR screening methods. The projects were heterogeneous with respect to search yield (median 9328; range 5243 to 47,385 records; interquartile range (IQR) 15,688 records), topic (Antipsychotics, Bronchiolitis, Diabetes, Child Health SRs), and screening complexity. We uploaded the records to Abstrackr and screened until it made predictions about the relevance of the remaining records. Across three trials for each project, we compared the predictions to human reviewer decisions and calculated the sensitivity, specificity, precision, false negative rate, proportion missed, and workload savings. Abstrackr's sensitivity was > 0.75 for all projects and the mean specificity ranged from 0.69 to 0.90 with the exception of Child Health SRs, for which it was 0.19. The precision (proportion of records correctly predicted as relevant) varied by screening task (median 26.6%; range 14.8 to 64.7%; IQR 29.7%). The median false negative rate (proportion of records incorrectly predicted as irrelevant) was 12.6% (range 3.5 to 21.2%; IQR 12.3%). The workload savings were often large (median 67.2%, range 9.5 to 88.4%; IQR 23.9%). The proportion missed (proportion of records predicted as irrelevant that were included in the final report, out of the total number predicted as irrelevant) was 0.1% for all SRs and 6.4% for the descriptive analysis. This equated to 4.2% (range 0 to 12.2%; IQR 7.8%) of the records in the final reports. Abstrackr's reliability and the workload savings varied by screening task. Workload savings came at the expense of potentially missing

  7. An Interactive Simulation System for Modeling Stands, Harvests, and Machines

    Treesearch

    Jingxin Wang; W. Dale Greene

    1999-01-01

    A interactive computer simulation program models stands, harvest, and machine factors and evaluates their interatcitons while performing felling, skidding, or fowarding activities. A stand generator allows the user to generate either natural or planted stands. Fellings with chainsaw, drive-to-tree feller-bunchers, or harvesters and extraction with grapple skidders or...

  8. An incremental anomaly detection model for virtual machines

    PubMed Central

    Zhang, Hancui; Chen, Shuyu; Liu, Jun; Zhou, Zhen; Wu, Tianshu

    2017-01-01

    Self-Organizing Map (SOM) algorithm as an unsupervised learning method has been applied in anomaly detection due to its capabilities of self-organizing and automatic anomaly prediction. However, because of the algorithm is initialized in random, it takes a long time to train a detection model. Besides, the Cloud platforms with large scale virtual machines are prone to performance anomalies due to their high dynamic and resource sharing characters, which makes the algorithm present a low accuracy and a low scalability. To address these problems, an Improved Incremental Self-Organizing Map (IISOM) model is proposed for anomaly detection of virtual machines. In this model, a heuristic-based initialization algorithm and a Weighted Euclidean Distance (WED) algorithm are introduced into SOM to speed up the training process and improve model quality. Meanwhile, a neighborhood-based searching algorithm is presented to accelerate the detection time by taking into account the large scale and high dynamic features of virtual machines on cloud platform. To demonstrate the effectiveness, experiments on a common benchmark KDD Cup dataset and a real dataset have been performed. Results suggest that IISOM has advantages in accuracy and convergence velocity of anomaly detection for virtual machines on cloud platform. PMID:29117245

  9. An incremental anomaly detection model for virtual machines.

    PubMed

    Zhang, Hancui; Chen, Shuyu; Liu, Jun; Zhou, Zhen; Wu, Tianshu

    2017-01-01

    Self-Organizing Map (SOM) algorithm as an unsupervised learning method has been applied in anomaly detection due to its capabilities of self-organizing and automatic anomaly prediction. However, because of the algorithm is initialized in random, it takes a long time to train a detection model. Besides, the Cloud platforms with large scale virtual machines are prone to performance anomalies due to their high dynamic and resource sharing characters, which makes the algorithm present a low accuracy and a low scalability. To address these problems, an Improved Incremental Self-Organizing Map (IISOM) model is proposed for anomaly detection of virtual machines. In this model, a heuristic-based initialization algorithm and a Weighted Euclidean Distance (WED) algorithm are introduced into SOM to speed up the training process and improve model quality. Meanwhile, a neighborhood-based searching algorithm is presented to accelerate the detection time by taking into account the large scale and high dynamic features of virtual machines on cloud platform. To demonstrate the effectiveness, experiments on a common benchmark KDD Cup dataset and a real dataset have been performed. Results suggest that IISOM has advantages in accuracy and convergence velocity of anomaly detection for virtual machines on cloud platform.

  10. Innovative model of business process reengineering at machine building enterprises

    NASA Astrophysics Data System (ADS)

    Nekrasov, R. Yu; Tempel, Yu A.; Tempel, O. A.

    2017-10-01

    The paper provides consideration of business process reengineering viewed as amanagerial innovation accepted by present day machine building enterprises, as well as waysto improve its procedure. A developed innovative model of reengineering measures isdescribed and is based on the process approach and other principles of company management.

  11. Control of discrete event systems modeled as hierarchical state machines

    NASA Technical Reports Server (NTRS)

    Brave, Y.; Heymann, M.

    1991-01-01

    The authors examine a class of discrete event systems (DESs) modeled as asynchronous hierarchical state machines (AHSMs). For this class of DESs, they provide an efficient method for testing reachability, which is an essential step in many control synthesis procedures. This method utilizes the asynchronous nature and hierarchical structure of AHSMs, thereby illustrating the advantage of the AHSM representation as compared with its equivalent (flat) state machine representation. An application of the method is presented where an online minimally restrictive solution is proposed for the problem of maintaining a controlled AHSM within prescribed legal bounds.

  12. Machine learning models in breast cancer survival prediction.

    PubMed

    Montazeri, Mitra; Montazeri, Mohadeseh; Montazeri, Mahdieh; Beigzadeh, Amin

    2016-01-01

    Breast cancer is one of the most common cancers with a high mortality rate among women. With the early diagnosis of breast cancer survival will increase from 56% to more than 86%. Therefore, an accurate and reliable system is necessary for the early diagnosis of this cancer. The proposed model is the combination of rules and different machine learning techniques. Machine learning models can help physicians to reduce the number of false decisions. They try to exploit patterns and relationships among a large number of cases and predict the outcome of a disease using historical cases stored in datasets. The objective of this study is to propose a rule-based classification method with machine learning techniques for the prediction of different types of Breast cancer survival. We use a dataset with eight attributes that include the records of 900 patients in which 876 patients (97.3%) and 24 (2.7%) patients were females and males respectively. Naive Bayes (NB), Trees Random Forest (TRF), 1-Nearest Neighbor (1NN), AdaBoost (AD), Support Vector Machine (SVM), RBF Network (RBFN), and Multilayer Perceptron (MLP) machine learning techniques with 10-cross fold technique were used with the proposed model for the prediction of breast cancer survival. The performance of machine learning techniques were evaluated with accuracy, precision, sensitivity, specificity, and area under ROC curve. Out of 900 patients, 803 patients and 97 patients were alive and dead, respectively. In this study, Trees Random Forest (TRF) technique showed better results in comparison to other techniques (NB, 1NN, AD, SVM and RBFN, MLP). The accuracy, sensitivity and the area under ROC curve of TRF are 96%, 96%, 93%, respectively. However, 1NN machine learning technique provided poor performance (accuracy 91%, sensitivity 91% and area under ROC curve 78%). This study demonstrates that Trees Random Forest model (TRF) which is a rule-based classification model was the best model with the highest level of

  13. Analytical model for force prediction when machining metal matrix composites

    NASA Astrophysics Data System (ADS)

    Sikder, Snahungshu

    Metal Matrix Composites (MMC) offer several thermo-mechanical advantages over standard materials and alloys which make them better candidates in different applications. Their light weight, high stiffness, and strength have attracted several industries such as automotive, aerospace, and defence for their wide range of products. However, the wide spread application of Meal Matrix Composites is still a challenge for industry. The hard and abrasive nature of the reinforcement particles is responsible for rapid tool wear and high machining costs. Fracture and debonding of the abrasive reinforcement particles are the considerable damage modes that directly influence the tool performance. It is very important to find highly effective way to machine MMCs. So, it is important to predict forces when machining Metal Matrix Composites because this will help to choose perfect tools for machining and ultimately save both money and time. This research presents an analytical force model for predicting the forces generated during machining of Metal Matrix Composites. In estimating the generated forces, several aspects of cutting mechanics were considered including: shearing force, ploughing force, and particle fracture force. Chip formation force was obtained by classical orthogonal metal cutting mechanics and the Johnson-Cook Equation. The ploughing force was formulated while the fracture force was calculated from the slip line field theory and the Griffith theory of failure. The predicted results were compared with previously measured data. The results showed very good agreement between the theoretically predicted and experimentally measured cutting forces.

  14. A comparative study of machine learning models for ethnicity classification

    NASA Astrophysics Data System (ADS)

    Trivedi, Advait; Bessie Amali, D. Geraldine

    2017-11-01

    This paper endeavours to adopt a machine learning approach to solve the problem of ethnicity recognition. Ethnicity identification is an important vision problem with its use cases being extended to various domains. Despite the multitude of complexity involved, ethnicity identification comes naturally to humans. This meta information can be leveraged to make several decisions, be it in target marketing or security. With the recent development of intelligent systems a sub module to efficiently capture ethnicity would be useful in several use cases. Several attempts to identify an ideal learning model to represent a multi-ethnic dataset have been recorded. A comparative study of classifiers such as support vector machines, logistic regression has been documented. Experimental results indicate that the logical classifier provides a much accurate classification than the support vector machine.

  15. Modeling Geomagnetic Variations using a Machine Learning Framework

    NASA Astrophysics Data System (ADS)

    Cheung, C. M. M.; Handmer, C.; Kosar, B.; Gerules, G.; Poduval, B.; Mackintosh, G.; Munoz-Jaramillo, A.; Bobra, M.; Hernandez, T.; McGranaghan, R. M.

    2017-12-01

    We present a framework for data-driven modeling of Heliophysics time series data. The Solar Terrestrial Interaction Neural net Generator (STING) is an open source python module built on top of state-of-the-art statistical learning frameworks (traditional machine learning methods as well as deep learning). To showcase the capability of STING, we deploy it for the problem of predicting the temporal variation of geomagnetic fields. The data used includes solar wind measurements from the OMNI database and geomagnetic field data taken by magnetometers at US Geological Survey observatories. We examine the predictive capability of different machine learning techniques (recurrent neural networks, support vector machines) for a range of forecasting times (minutes to 12 hours). STING is designed to be extensible to other types of data. We show how STING can be used on large sets of data from different sensors/observatories and adapted to tackle other problems in Heliophysics.

  16. Machine Visual Targeting Modeled on Biological Reflexes

    DTIC Science & Technology

    1993-02-01

    Released by Under authority of P. J. Heckman, Jr., Head N. B . Estabrook Undersea Al and Robotics Branch Ocean Engineering Division ACKNOWLEDGMENTS...elements are modeled after the small B type described by Sterling (1983) that have no gap junctions, and synapse only wvith rods and rod bipolars. Both on...of Y retina for motion detection showing receptors (R), horizonU~L% (H). on-center bilwobrs ( B ] . off-center bipolars (00), arnacrine (A). on-center

  17. Thermal-mechanical modeling of laser ablation hybrid machining

    NASA Astrophysics Data System (ADS)

    Matin, Mohammad Kaiser

    2001-08-01

    Hard, brittle and wear-resistant materials like ceramics pose a problem when being machined using conventional machining processes. Machining ceramics even with a diamond cutting tool is very difficult and costly. Near net-shape processes, like laser evaporation, produce micro-cracks that require extra finishing. Thus it is anticipated that ceramic machining will have to continue to be explored with new-sprung techniques before ceramic materials become commonplace. This numerical investigation results from the numerical simulations of the thermal and mechanical modeling of simultaneous material removal from hard-to-machine materials using both laser ablation and conventional tool cutting utilizing the finite element method. The model is formulated using a two dimensional, planar, computational domain. The process simulation acronymed, LAHM (Laser Ablation Hybrid Machining), uses laser energy for two purposes. The first purpose is to remove the material by ablation. The second purpose is to heat the unremoved material that lies below the ablated material in order to ``soften'' it. The softened material is then simultaneously removed by conventional machining processes. The complete solution determines the temperature distribution and stress contours within the material and tracks the moving boundary that occurs due to material ablation. The temperature distribution is used to determine the distance below the phase change surface where sufficient ``softening'' has occurred, so that a cutting tool may be used to remove additional material. The model incorporated for tracking the ablative surface does not assume an isothermal melt phase (e.g. Stefan problem) for laser ablation. Both surface absorption and volume absorption of laser energy as function of depth have been considered in the models. LAHM, from the thermal and mechanical point of view is a complex machining process involving large deformations at high strain rates, thermal effects of the laser, removal of

  18. Machine learning modelling for predicting soil liquefaction susceptibility

    NASA Astrophysics Data System (ADS)

    Samui, P.; Sitharam, T. G.

    2011-01-01

    This study describes two machine learning techniques applied to predict liquefaction susceptibility of soil based on the standard penetration test (SPT) data from the 1999 Chi-Chi, Taiwan earthquake. The first machine learning technique which uses Artificial Neural Network (ANN) based on multi-layer perceptions (MLP) that are trained with Levenberg-Marquardt backpropagation algorithm. The second machine learning technique uses the Support Vector machine (SVM) that is firmly based on the theory of statistical learning theory, uses classification technique. ANN and SVM have been developed to predict liquefaction susceptibility using corrected SPT [(N1)60] and cyclic stress ratio (CSR). Further, an attempt has been made to simplify the models, requiring only the two parameters [(N1)60 and peck ground acceleration (amax/g)], for the prediction of liquefaction susceptibility. The developed ANN and SVM models have also been applied to different case histories available globally. The paper also highlights the capability of the SVM over the ANN models.

  19. Aspheric glass lens modeling and machining

    NASA Astrophysics Data System (ADS)

    Johnson, R. Barry; Mandina, Michael

    2005-08-01

    The incorporation of aspheric lenses in complex lens system can provide significant image quality improvement, reduction of the number of lens elements, smaller size, and lower weight. Recently, it has become practical to manufacture aspheric glass lenses using diamond-grinding methods. The evolution of the manufacturing technology is discussed for a specific aspheric glass lens. When a prototype all-glass lens system (80 mm efl, F/2.5) was fabricated and tested, it was observed that the image quality was significantly less than was predicted by the optical design software. The cause of the degradation was identified as the large aspheric element in the lens. Identification was possible by precision mapping of the spatial coordinates of the lens surface and then transforming this data into an appropriate optical surface defined by derived grid sag data. The resulting optical analysis yielded a modeled image consistent with that observed when testing the prototype lens system in the laboratory. This insight into a localized slope-error problem allowed improvements in the fabrication process to be implemented. The second fabrication attempt, the resulting aspheric lens provided remarkable improvement in the observed image quality, although still falling somewhat short of the desired image quality goal. In parallel with the fabrication enhancement effort, optical modeling of the surface was undertaken to determine how much surface error and error types were allowable to achieve the desired image quality goal. With this knowledge, final improvements were made to the fabrication process. The third prototype lens achieved the goal of optical performance. Rapid development of the aspheric glass lens was made possible by the interactive relationship between the optical designer, diamond-grinding personnel, and the metrology personnel. With rare exceptions, the subsequent production lenses were optical acceptable and afforded reasonable manufacturing costs.

  20. Problems in modeling man machine control behavior in biodynamic environments

    NASA Technical Reports Server (NTRS)

    Jex, H. R.

    1972-01-01

    Reviewed are some current problems in modeling man-machine control behavior in a biodynamic environment. It is given in two parts: (1) a review of the models which are appropriate for manual control behavior and the added elements necessary to deal with biodynamic interfaces; and (2) a review of some biodynamic interface pilot/vehicle problems which have occurred, been solved, or need to be solved.

  1. The rise of machine consciousness: studying consciousness with computational models.

    PubMed

    Reggia, James A

    2013-08-01

    Efforts to create computational models of consciousness have accelerated over the last two decades, creating a field that has become known as artificial consciousness. There have been two main motivations for this controversial work: to develop a better scientific understanding of the nature of human/animal consciousness and to produce machines that genuinely exhibit conscious awareness. This review begins by briefly explaining some of the concepts and terminology used by investigators working on machine consciousness, and summarizes key neurobiological correlates of human consciousness that are particularly relevant to past computational studies. Models of consciousness developed over the last twenty years are then surveyed. These models are largely found to fall into five categories based on the fundamental issue that their developers have selected as being most central to consciousness: a global workspace, information integration, an internal self-model, higher-level representations, or attention mechanisms. For each of these five categories, an overview of past work is given, a representative example is presented in some detail to illustrate the approach, and comments are provided on the contributions and limitations of the methodology. Three conclusions are offered about the state of the field based on this review: (1) computational modeling has become an effective and accepted methodology for the scientific study of consciousness, (2) existing computational models have successfully captured a number of neurobiological, cognitive, and behavioral correlates of conscious information processing as machine simulations, and (3) no existing approach to artificial consciousness has presented a compelling demonstration of phenomenal machine consciousness, or even clear evidence that artificial phenomenal consciousness will eventually be possible. The paper concludes by discussing the importance of continuing work in this area, considering the ethical issues it raises

  2. Development of machine learning models for diagnosis of glaucoma.

    PubMed

    Kim, Seong Jae; Cho, Kyong Jin; Oh, Sejong

    2017-01-01

    The study aimed to develop machine learning models that have strong prediction power and interpretability for diagnosis of glaucoma based on retinal nerve fiber layer (RNFL) thickness and visual field (VF). We collected various candidate features from the examination of retinal nerve fiber layer (RNFL) thickness and visual field (VF). We also developed synthesized features from original features. We then selected the best features proper for classification (diagnosis) through feature evaluation. We used 100 cases of data as a test dataset and 399 cases of data as a training and validation dataset. To develop the glaucoma prediction model, we considered four machine learning algorithms: C5.0, random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN). We repeatedly composed a learning model using the training dataset and evaluated it by using the validation dataset. Finally, we got the best learning model that produces the highest validation accuracy. We analyzed quality of the models using several measures. The random forest model shows best performance and C5.0, SVM, and KNN models show similar accuracy. In the random forest model, the classification accuracy is 0.98, sensitivity is 0.983, specificity is 0.975, and AUC is 0.979. The developed prediction models show high accuracy, sensitivity, specificity, and AUC in classifying among glaucoma and healthy eyes. It will be used for predicting glaucoma against unknown examination records. Clinicians may reference the prediction results and be able to make better decisions. We may combine multiple learning models to increase prediction accuracy. The C5.0 model includes decision rules for prediction. It can be used to explain the reasons for specific predictions.

  3. Abstract Model of the SATS Concept of Operations: Initial Results and Recommendations

    NASA Technical Reports Server (NTRS)

    Dowek, Gilles; Munoz, Cesar; Carreno, Victor A.

    2004-01-01

    An abstract mathematical model of the concept of operations for the Small Aircraft Transportation System (SATS) is presented. The Concept of Operations consist of several procedures that describe nominal operations for SATS, Several safety properties of the system are proven using formal techniques. The final goal of the verification effort is to show that under nominal operations, aircraft are safely separated. The abstract model was written and formally verified in the Prototype Verification System (PVS).

  4. 97. View of International Business Machine (IBM) digital computer model ...

    Library of Congress Historic Buildings Survey, Historic Engineering Record, Historic Landscapes Survey

    97. View of International Business Machine (IBM) digital computer model 7090 magnetic core installation, international telephone and telegraph (ITT) Artic Services Inc., Official photograph BMEWS site II, Clear, AK, by unknown photographer, 17 September 1965, BMEWS, clear as negative no. A-6604. - Clear Air Force Station, Ballistic Missile Early Warning System Site II, One mile west of mile marker 293.5 on Parks Highway, 5 miles southwest of Anderson, Anderson, Denali Borough, AK

  5. Tracer transport in soils and shallow groundwater: model abstraction with modern tools

    USDA-ARS?s Scientific Manuscript database

    Vadose zone controls contaminant transport from the surface to groundwater, and modeling transport in vadose zone has become a burgeoning field. Exceedingly complex models of subsurface contaminant transport are often inefficient. Model abstraction is the methodology for reducing the complexity of a...

  6. Machine Learning Predictions of a Multiresolution Climate Model Ensemble

    NASA Astrophysics Data System (ADS)

    Anderson, Gemma J.; Lucas, Donald D.

    2018-05-01

    Statistical models of high-resolution climate models are useful for many purposes, including sensitivity and uncertainty analyses, but building them can be computationally prohibitive. We generated a unique multiresolution perturbed parameter ensemble of a global climate model. We use a novel application of a machine learning technique known as random forests to train a statistical model on the ensemble to make high-resolution model predictions of two important quantities: global mean top-of-atmosphere energy flux and precipitation. The random forests leverage cheaper low-resolution simulations, greatly reducing the number of high-resolution simulations required to train the statistical model. We demonstrate that high-resolution predictions of these quantities can be obtained by training on an ensemble that includes only a small number of high-resolution simulations. We also find that global annually averaged precipitation is more sensitive to resolution changes than to any of the model parameters considered.

  7. Editorial: Mathematical Methods and Modeling in Machine Fault Diagnosis

    DOE PAGES

    Yan, Ruqiang; Chen, Xuefeng; Li, Weihua; ...

    2014-12-18

    Modern mathematics has commonly been utilized as an effective tool to model mechanical equipment so that their dynamic characteristics can be studied analytically. This will help identify potential failures of mechanical equipment by observing change in the equipment’s dynamic parameters. On the other hand, dynamic signals are also important and provide reliable information about the equipment’s working status. Modern mathematics has also provided us with a systematic way to design and implement various signal processing methods, which are used to analyze these dynamic signals, and to enhance intrinsic signal components that are directly related to machine failures. This special issuemore » is aimed at stimulating not only new insights on mathematical methods for modeling but also recently developed signal processing methods, such as sparse decomposition with potential applications in machine fault diagnosis. Finally, the papers included in this special issue provide a glimpse into some of the research and applications in the field of machine fault diagnosis through applications of the modern mathematical methods.« less

  8. Modeling Music Emotion Judgments Using Machine Learning Methods

    PubMed Central

    Vempala, Naresh N.; Russo, Frank A.

    2018-01-01

    Emotion judgments and five channels of physiological data were obtained from 60 participants listening to 60 music excerpts. Various machine learning (ML) methods were used to model the emotion judgments inclusive of neural networks, linear regression, and random forests. Input for models of perceived emotion consisted of audio features extracted from the music recordings. Input for models of felt emotion consisted of physiological features extracted from the physiological recordings. Models were trained and interpreted with consideration of the classic debate in music emotion between cognitivists and emotivists. Our models supported a hybrid position wherein emotion judgments were influenced by a combination of perceived and felt emotions. In comparing the different ML approaches that were used for modeling, we conclude that neural networks were optimal, yielding models that were flexible as well as interpretable. Inspection of a committee machine, encompassing an ensemble of networks, revealed that arousal judgments were predominantly influenced by felt emotion, whereas valence judgments were predominantly influenced by perceived emotion. PMID:29354080

  9. Model-Driven Engineering of Machine Executable Code

    NASA Astrophysics Data System (ADS)

    Eichberg, Michael; Monperrus, Martin; Kloppenburg, Sven; Mezini, Mira

    Implementing static analyses of machine-level executable code is labor intensive and complex. We show how to leverage model-driven engineering to facilitate the design and implementation of programs doing static analyses. Further, we report on important lessons learned on the benefits and drawbacks while using the following technologies: using the Scala programming language as target of code generation, using XML-Schema to express a metamodel, and using XSLT to implement (a) transformations and (b) a lint like tool. Finally, we report on the use of Prolog for writing model transformations.

  10. Modeling of Passive Forces of Machine Tool Covers

    NASA Astrophysics Data System (ADS)

    Kolar, Petr; Hudec, Jan; Sulitka, Matej

    The passive forces acting against the drive force are phenomena that influence dynamical properties and precision of linear axes equipped with feed drives. Covers are one of important sources of passive forces in machine tools. The paper describes virtual evaluation of cover passive forces using the cover complex model. The model is able to compute interaction between flexible cover segments and sealing wiper. The result is deformation of cover segments and wipers which is used together with measured friction coefficient for computation of cover total passive force. This resulting passive force is dependent on cover position. Comparison of computational results and measurement on the real cover is presented in the paper.

  11. "Machine" consciousness and "artificial" thought: an operational architectonics model guided approach.

    PubMed

    Fingelkurts, Andrew A; Fingelkurts, Alexander A; Neves, Carlos F H

    2012-01-05

    Instead of using low-level neurophysiology mimicking and exploratory programming methods commonly used in the machine consciousness field, the hierarchical operational architectonics (OA) framework of brain and mind functioning proposes an alternative conceptual-theoretical framework as a new direction in the area of model-driven machine (robot) consciousness engineering. The unified brain-mind theoretical OA model explicitly captures (though in an informal way) the basic essence of brain functional architecture, which indeed constitutes a theory of consciousness. The OA describes the neurophysiological basis of the phenomenal level of brain organization. In this context the problem of producing man-made "machine" consciousness and "artificial" thought is a matter of duplicating all levels of the operational architectonics hierarchy (with its inherent rules and mechanisms) found in the brain electromagnetic field. We hope that the conceptual-theoretical framework described in this paper will stimulate the interest of mathematicians and/or computer scientists to abstract and formalize principles of hierarchy of brain operations which are the building blocks for phenomenal consciousness and thought. Copyright © 2010 Elsevier B.V. All rights reserved.

  12. Calibrating Building Energy Models Using Supercomputer Trained Machine Learning Agents

    SciTech Connect

    Sanyal, Jibonananda; New, Joshua Ryan; Edwards, Richard

    2014-01-01

    Building Energy Modeling (BEM) is an approach to model the energy usage in buildings for design and retrofit purposes. EnergyPlus is the flagship Department of Energy software that performs BEM for different types of buildings. The input to EnergyPlus can often extend in the order of a few thousand parameters which have to be calibrated manually by an expert for realistic energy modeling. This makes it challenging and expensive thereby making building energy modeling unfeasible for smaller projects. In this paper, we describe the Autotune research which employs machine learning algorithms to generate agents for the different kinds of standardmore » reference buildings in the U.S. building stock. The parametric space and the variety of building locations and types make this a challenging computational problem necessitating the use of supercomputers. Millions of EnergyPlus simulations are run on supercomputers which are subsequently used to train machine learning algorithms to generate agents. These agents, once created, can then run in a fraction of the time thereby allowing cost-effective calibration of building models.« less

  13. Machine learning and docking models for Mycobacterium tuberculosis topoisomerase I.

    PubMed

    Ekins, Sean; Godbole, Adwait Anand; Kéri, György; Orfi, Lászlo; Pato, János; Bhat, Rajeshwari Subray; Verma, Rinkee; Bradley, Erin K; Nagaraja, Valakunja

    2017-03-01

    There is a shortage of compounds that are directed towards new targets apart from those targeted by the FDA approved drugs used against Mycobacterium tuberculosis. Topoisomerase I (Mttopo I) is an essential mycobacterial enzyme and a promising target in this regard. However, it suffers from a shortage of known inhibitors. We have previously used computational approaches such as homology modeling and docking to propose 38 FDA approved drugs for testing and identified several active molecules. To follow on from this, we now describe the in vitro testing of a library of 639 compounds. These data were used to create machine learning models for Mttopo I which were further validated. The combined Mttopo I Bayesian model had a 5 fold cross validation receiver operator characteristic of 0.74 and sensitivity, specificity and concordance values above 0.76 and was used to select commercially available compounds for testing in vitro. The recently described crystal structure of Mttopo I was also compared with the previously described homology model and then used to dock the Mttopo I actives norclomipramine and imipramine. In summary, we describe our efforts to identify small molecule inhibitors of Mttopo I using a combination of machine learning modeling and docking studies in conjunction with screening of the selected molecules for enzyme inhibition. We demonstrate the experimental inhibition of Mttopo I by small molecule inhibitors and show that the enzyme can be readily targeted for lead molecule development. Copyright © 2017 Elsevier Ltd. All rights reserved.

  14. Assessing the impacts of water abstractions on river ecosystem services: an eco-hydraulic modelling approach

    SciTech Connect

    Carolli, Mauro, E-mail: mauro.carolli@unitn.it; Geneletti, Davide, E-mail: davide.geneletti@unitn.it; Zolezzi, Guido, E-mail: guido.zolezzi@unitn.it

    The provision of important river ecosystem services (ES) is dependent on the flow regime. This requires methods to assess the impacts on ES caused by interventions on rivers that affect flow regime, such as water abstractions. This study proposes a method to i) quantify the provision of a set of river ES, ii) simulate the effects of water abstraction alternatives that differ in location and abstracted flow, and iii) assess the impact of water abstraction alternatives on the selected ES. The method is based on river modelling science, and integrates spatially distributed hydrological, hydraulic and habitat models at different spatialmore » and temporal scales. The method is applied to the hydropeaked upper Noce River (Northern Italy), which is regulated by hydropower operations. We selected locally relevant river ES: habitat suitability for the adult marble trout, white-water rafting suitability, hydroelectricity production from run-of-river (RoR) plants. Our results quantify the seasonality of river ES response variables and their intrinsic non-linearity, which explains why the same abstracted flow can produce different effects on trout habitat and rafting suitability depending on the morphology of the abstracted reach. An economic valuation of the examined river ES suggests that incomes from RoR hydropower plants are of comparable magnitude to touristic revenue losses related to the decrease in rafting suitability.« less

  15. Machine learning models for lipophilicity and their domain of applicability.

    PubMed

    Schroeter, Timon; Schwaighofer, Anton; Mika, Sebastian; Laak, Antonius Ter; Suelzle, Detlev; Ganzer, Ursula; Heinrich, Nikolaus; Müller, Klaus-Robert

    2007-01-01

    Unfavorable lipophilicity and water solubility cause many drug failures; therefore these properties have to be taken into account early on in lead discovery. Commercial tools for predicting lipophilicity usually have been trained on small and neutral molecules, and are thus often unable to accurately predict in-house data. Using a modern Bayesian machine learning algorithm--a Gaussian process model--this study constructs a log D7 model based on 14,556 drug discovery compounds of Bayer Schering Pharma. Performance is compared with support vector machines, decision trees, ridge regression, and four commercial tools. In a blind test on 7013 new measurements from the last months (including compounds from new projects) 81% were predicted correctly within 1 log unit, compared to only 44% achieved by commercial software. Additional evaluations using public data are presented. We consider error bars for each method (model based error bars, ensemble based, and distance based approaches), and investigate how well they quantify the domain of applicability of each model.

  16. Machine Learning Techniques for Global Sensitivity Analysis in Climate Models

    NASA Astrophysics Data System (ADS)

    Safta, C.; Sargsyan, K.; Ricciuto, D. M.

    2017-12-01

    Climate models studies are not only challenged by the compute intensive nature of these models but also by the high-dimensionality of the input parameter space. In our previous work with the land model components (Sargsyan et al., 2014) we identified subsets of 10 to 20 parameters relevant for each QoI via Bayesian compressive sensing and variance-based decomposition. Nevertheless the algorithms were challenged by the nonlinear input-output dependencies for some of the relevant QoIs. In this work we will explore a combination of techniques to extract relevant parameters for each QoI and subsequently construct surrogate models with quantified uncertainty necessary to future developments, e.g. model calibration and prediction studies. In the first step, we will compare the skill of machine-learning models (e.g. neural networks, support vector machine) to identify the optimal number of classes in selected QoIs and construct robust multi-class classifiers that will partition the parameter space in regions with smooth input-output dependencies. These classifiers will be coupled with techniques aimed at building sparse and/or low-rank surrogate models tailored to each class. Specifically we will explore and compare sparse learning techniques with low-rank tensor decompositions. These models will be used to identify parameters that are important for each QoI. Surrogate accuracy requirements are higher for subsequent model calibration studies and we will ascertain the performance of this workflow for multi-site ALM simulation ensembles.

  17. Using machine learning to model dose-response relationships.

    PubMed

    Linden, Ariel; Yarnold, Paul R; Nallamothu, Brahmajee K

    2016-12-01

    Establishing the relationship between various doses of an exposure and a response variable is integral to many studies in health care. Linear parametric models, widely used for estimating dose-response relationships, have several limitations. This paper employs the optimal discriminant analysis (ODA) machine-learning algorithm to determine the degree to which exposure dose can be distinguished based on the distribution of the response variable. By framing the dose-response relationship as a classification problem, machine learning can provide the same functionality as conventional models, but can additionally make individual-level predictions, which may be helpful in practical applications like establishing responsiveness to prescribed drug regimens. Using data from a study measuring the responses of blood flow in the forearm to the intra-arterial administration of isoproterenol (separately for 9 black and 13 white men, and pooled), we compare the results estimated from a generalized estimating equations (GEE) model with those estimated using ODA. Generalized estimating equations and ODA both identified many statistically significant dose-response relationships, separately by race and for pooled data. Post hoc comparisons between doses indicated ODA (based on exact P values) was consistently more conservative than GEE (based on estimated P values). Compared with ODA, GEE produced twice as many instances of paradoxical confounding (findings from analysis of pooled data that are inconsistent with findings from analyses stratified by race). Given its unique advantages and greater analytic flexibility, maximum-accuracy machine-learning methods like ODA should be considered as the primary analytic approach in dose-response applications. © 2016 John Wiley & Sons, Ltd.

  18. Selected translated abstracts of Russian-language climate-change publications. 4: General circulation models (in English;Russian)

    SciTech Connect

    Burtis, M.D.; Razuvaev, V.N.; Sivachok, S.G.

    1996-10-01

    This report presents English-translated abstracts of important Russian-language literature concerning general circulation models as they relate to climate change. Into addition to the bibliographic citations and abstracts translated into English, this report presents the original citations and abstracts in Russian. Author and title indexes are included to assist the reader in locating abstracts of particular interest.

  19. Abstraction of an Affective-Cognitive Decision Making Model Based on Simulated Behaviour and Perception Chains

    NASA Astrophysics Data System (ADS)

    Sharpanskykh, Alexei; Treur, Jan

    Employing rich internal agent models of actors in large-scale socio-technical systems often results in scalability issues. The problem addressed in this paper is how to improve computational properties of a complex internal agent model, while preserving its behavioral properties. The problem is addressed for the case of an existing affective-cognitive decision making model instantiated for an emergency scenario. For this internal decision model an abstracted behavioral agent model is obtained, which ensures a substantial increase of the computational efficiency at the cost of approximately 1% behavioural error. The abstraction technique used can be applied to a wide range of internal agent models with loops, for example, involving mutual affective-cognitive interactions.

  20. Advanced Machine Learning Emulators of Radiative Transfer Models

    NASA Astrophysics Data System (ADS)

    Camps-Valls, G.; Verrelst, J.; Martino, L.; Vicent, J.

    2017-12-01

    Physically-based model inversion methodologies are based on physical laws and established cause-effect relationships. A plethora of remote sensing applications rely on the physical inversion of a Radiative Transfer Model (RTM), which lead to physically meaningful bio-geo-physical parameter estimates. The process is however computationally expensive, needs expert knowledge for both the selection of the RTM, its parametrization and the the look-up table generation, as well as its inversion. Mimicking complex codes with statistical nonlinear machine learning algorithms has become the natural alternative very recently. Emulators are statistical constructs able to approximate the RTM, although at a fraction of the computational cost, providing an estimation of uncertainty, and estimations of the gradient or finite integral forms. We review the field and recent advances of emulation of RTMs with machine learning models. We posit Gaussian processes (GPs) as the proper framework to tackle the problem. Furthermore, we introduce an automatic methodology to construct emulators for costly RTMs. The Automatic Gaussian Process Emulator (AGAPE) methodology combines the interpolation capabilities of GPs with the accurate design of an acquisition function that favours sampling in low density regions and flatness of the interpolation function. We illustrate the good capabilities of our emulators in toy examples, leaf and canopy levels PROSPECT and PROSAIL RTMs, and for the construction of an optimal look-up-table for atmospheric correction based on MODTRAN5.

  1. Modeling of Unsteady Three-dimensional Flows in Multistage Machines

    NASA Technical Reports Server (NTRS)

    Hall, Kenneth C.; Pratt, Edmund T., Jr.; Kurkov, Anatole (Technical Monitor)

    2003-01-01

    Despite many years of development, the accurate and reliable prediction of unsteady aerodynamic forces acting on turbomachinery blades remains less than satisfactory, especially when viewed next to the great success investigators have had in predicting steady flows. Hall and Silkowski (1997) have proposed that one of the main reasons for the discrepancy between theory and experiment and/or industrial experience is that many of the current unsteady aerodynamic theories model a single blade row in an infinitely long duct, ignoring potentially important multistage effects. However, unsteady flows are made up of acoustic, vortical, and entropic waves. These waves provide a mechanism for the rotors and stators of multistage machines to communicate with one another. In other words, wave behavior makes unsteady flows fundamentally a multistage (and three-dimensional) phenomenon. In this research program, we have has as goals (1) the development of computationally efficient computer models of the unsteady aerodynamic response of blade rows embedded in a multistage machine (these models will ultimately be capable of analyzing three-dimensional viscous transonic flows), and (2) the use of these computer codes to study a number of important multistage phenomena.

  2. Process Approach for Modeling of Machine and Tractor Fleet Structure

    NASA Astrophysics Data System (ADS)

    Dokin, B. D.; Aletdinova, A. A.; Kravchenko, M. S.; Tsybina, Y. S.

    2018-05-01

    The existing software complexes on modelling of the machine and tractor fleet structure are mostly aimed at solving the task of optimization. However, the creators, choosing only one optimization criterion and incorporating it in their software, provide grounds on why it is the best without giving a decision maker the opportunity to choose it for their enterprise. To analyze “bottlenecks” of machine and tractor fleet modelling, the authors of this article created a process model, in which they included adjustment to the plan of using machinery based on searching through alternative technologies. As a result, the following recommendations for software complex development have been worked out: the introduction of a database of alternative technologies; the possibility for a user to change the timing of the operations even beyond the allowable limits and in that case the calculation of the incurred loss; the possibility to rule out the solution of an optimization task, and if there is a necessity in it - the possibility to choose an optimization criterion; introducing graphical display of an annual complex of works, which could be enough for the development and adjustment of a business strategy.

  3. Modeling the Swift BAT Trigger Algorithm with Machine Learning

    NASA Technical Reports Server (NTRS)

    Graff, Philip B.; Lien, Amy Y.; Baker, John G.; Sakamoto, Takanori

    2015-01-01

    To draw inferences about gamma-ray burst (GRB) source populations based on Swift observations, it is essential to understand the detection efficiency of the Swift burst alert telescope (BAT). This study considers the problem of modeling the Swift BAT triggering algorithm for long GRBs, a computationally expensive procedure, and models it using machine learning algorithms. A large sample of simulated GRBs from Lien et al. (2014) is used to train various models: random forests, boosted decision trees (with AdaBoost), support vector machines, and artificial neural networks. The best models have accuracies of approximately greater than 97% (approximately less than 3% error), which is a significant improvement on a cut in GRB flux which has an accuracy of 89:6% (10:4% error). These models are then used to measure the detection efficiency of Swift as a function of redshift z, which is used to perform Bayesian parameter estimation on the GRB rate distribution. We find a local GRB rate density of eta(sub 0) approximately 0.48(+0.41/-0.23) Gpc(exp -3) yr(exp -1) with power-law indices of eta(sub 1) approximately 1.7(+0.6/-0.5) and eta(sub 2) approximately -5.9(+5.7/-0.1) for GRBs above and below a break point of z(sub 1) approximately 6.8(+2.8/-3.2). This methodology is able to improve upon earlier studies by more accurately modeling Swift detection and using this for fully Bayesian model fitting. The code used in this is analysis is publicly available online.

  4. Modeling the Swift Bat Trigger Algorithm with Machine Learning

    NASA Technical Reports Server (NTRS)

    Graff, Philip B.; Lien, Amy Y.; Baker, John G.; Sakamoto, Takanori

    2016-01-01

    To draw inferences about gamma-ray burst (GRB) source populations based on Swift observations, it is essential to understand the detection efficiency of the Swift burst alert telescope (BAT). This study considers the problem of modeling the Swift / BAT triggering algorithm for long GRBs, a computationally expensive procedure, and models it using machine learning algorithms. A large sample of simulated GRBs from Lien et al. is used to train various models: random forests, boosted decision trees (with AdaBoost), support vector machines, and artificial neural networks. The best models have accuracies of greater than or equal to 97 percent (less than or equal to 3 percent error), which is a significant improvement on a cut in GRB flux, which has an accuracy of 89.6 percent (10.4 percent error). These models are then used to measure the detection efficiency of Swift as a function of redshift z, which is used to perform Bayesian parameter estimation on the GRB rate distribution. We find a local GRB rate density of n (sub 0) approaching 0.48 (sup plus 0.41) (sub minus 0.23) per cubic gigaparsecs per year with power-law indices of n (sub 1) approaching 1.7 (sup plus 0.6) (sub minus 0.5) and n (sub 2) approaching minus 5.9 (sup plus 5.7) (sub minus 0.1) for GRBs above and below a break point of z (redshift) (sub 1) approaching 6.8 (sup plus 2.8) (sub minus 3.2). This methodology is able to improve upon earlier studies by more accurately modeling Swift detection and using this for fully Bayesian model fitting.

  5. Ecological footprint model using the support vector machine technique.

    PubMed

    Ma, Haibo; Chang, Wenjuan; Cui, Guangbai

    2012-01-01

    The per capita ecological footprint (EF) is one of the most widely recognized measures of environmental sustainability. It aims to quantify the Earth's biological resources required to support human activity. In this paper, we summarize relevant previous literature, and present five factors that influence per capita EF. These factors are: National gross domestic product (GDP), urbanization (independent of economic development), distribution of income (measured by the Gini coefficient), export dependence (measured by the percentage of exports to total GDP), and service intensity (measured by the percentage of service to total GDP). A new ecological footprint model based on a support vector machine (SVM), which is a machine-learning method based on the structural risk minimization principle from statistical learning theory was conducted to calculate the per capita EF of 24 nations using data from 123 nations. The calculation accuracy was measured by average absolute error and average relative error. They were 0.004883 and 0.351078% respectively. Our results demonstrate that the EF model based on SVM has good calculation performance.

  6. Hidden physics models: Machine learning of nonlinear partial differential equations

    NASA Astrophysics Data System (ADS)

    Raissi, Maziar; Karniadakis, George Em

    2018-03-01

    While there is currently a lot of enthusiasm about "big data", useful data is usually "small" and expensive to acquire. In this paper, we present a new paradigm of learning partial differential equations from small data. In particular, we introduce hidden physics models, which are essentially data-efficient learning machines capable of leveraging the underlying laws of physics, expressed by time dependent and nonlinear partial differential equations, to extract patterns from high-dimensional data generated from experiments. The proposed methodology may be applied to the problem of learning, system identification, or data-driven discovery of partial differential equations. Our framework relies on Gaussian processes, a powerful tool for probabilistic inference over functions, that enables us to strike a balance between model complexity and data fitting. The effectiveness of the proposed approach is demonstrated through a variety of canonical problems, spanning a number of scientific domains, including the Navier-Stokes, Schrödinger, Kuramoto-Sivashinsky, and time dependent linear fractional equations. The methodology provides a promising new direction for harnessing the long-standing developments of classical methods in applied mathematics and mathematical physics to design learning machines with the ability to operate in complex domains without requiring large quantities of data.

  7. Modelling fate and transport of pesticides in river catchments with drinking water abstractions

    NASA Astrophysics Data System (ADS)

    Desmet, Nele; Seuntjens, Piet; Touchant, Kaatje

    2010-05-01

    When drinking water is abstracted from surface water, the presence of pesticides may have a large impact on the purification costs. In order to respect imposed thresholds at points of drinking water abstraction in a river catchment, sustainable pesticide management strategies might be required in certain areas. To improve management strategies, a sound understanding of the emission routes, the transport, the environmental fate and the sources of pesticides is needed. However, pesticide monitoring data on which measures are founded, are generally scarce. Data scarcity hampers the interpretation and the decision making. In such a case, a modelling approach can be very useful as a tool to obtain complementary information. Modelling allows to take into account temporal and spatial variability in both discharges and concentrations. In the Netherlands, the Meuse river is used for drinking water abstraction and the government imposes the European drinking water standard for individual pesticides (0.1 ?g.L-1) for surface waters at points of drinking water abstraction. The reported glyphosate concentrations in the Meuse river frequently exceed the standard and this enhances the request for targeted measures. In this study, a model for the Meuse river was developed to estimate the contribution of influxes at the Dutch-Belgian border on the concentration levels detected at the drinking water intake 250 km downstream and to assess the contribution of the tributaries to the glyphosate loads. The effects of glyphosate decay on environmental fate were considered as well. Our results show that the application of a river model allows to asses fate and transport of pesticides in a catchment in spite of monitoring data scarcity. Furthermore, the model provides insight in the contribution of different sub basins to the pollution level. The modelling results indicate that the effect of local measures to reduce pesticides concentrations in the river at points of drinking water

  8. Labyrinth, An Abstract Model for Hypermedia Applications. Description of its Static Components.

    ERIC Educational Resources Information Center

    Diaz, Paloma; Aedo, Ignacio; Panetsos, Fivos

    1997-01-01

    The model for hypermedia applications called Labyrinth allows: (1) the design of platform-independent hypermedia applications; (2) the categorization, generalization and abstraction of sparse unstructured heterogeneous information in multiple and interconnected levels; (3) the creation of personal views in multiuser hyperdocuments for both groups…

  9. The Abstract Selection Task: New Data and an Almost Comprehensive Model

    ERIC Educational Resources Information Center

    Klauer, Karl Christoph; Stahl, Christoph; Erdfelder, Edgar

    2007-01-01

    A complete quantitative account of P. Wason's (1966) abstract selection task is proposed. The account takes the form of a mathematical model. It is assumed that some response patterns are caused by inferential reasoning, whereas other responses reflect cognitive processes that affect each card selection separately and independently of other card…

  10. Vehicle Concept Model Abstractions For Integrated Geometric, Inertial ,Rigid Body, Powertrain and FE Analysis

    DTIC Science & Technology

    2011-06-17

    structure through quantitative assessment of stiffness and modal parameter changes resulting from modifications to the beam geometries and positions...power transmission assembly. If the power limit at a wheel exceeds the traction limit, then depending on the type of differential placed on the axle ...components with appropriate model connectivity instead to determine the free modal response of powertrain type components, without abstraction

  11. Contention Modeling for Multithreaded Distributed Shared Memory Machines: The Cray XMT

    SciTech Connect

    Secchi, Simone; Tumeo, Antonino; Villa, Oreste

    Distributed Shared Memory (DSM) machines are a wide class of multi-processor computing systems where a large virtually-shared address space is mapped on a network of physically distributed memories. High memory latency and network contention are two of the main factors that limit performance scaling of such architectures. Modern high-performance computing DSM systems have evolved toward exploitation of massive hardware multi-threading and fine-grained memory hashing to tolerate irregular latencies, avoid network hot-spots and enable high scaling. In order to model the performance of such large-scale machines, parallel simulation has been proved to be a promising approach to achieve good accuracy inmore » reasonable times. One of the most critical factors in solving the simulation speed-accuracy trade-off is network modeling. The Cray XMT is a massively multi-threaded supercomputing architecture that belongs to the DSM class, since it implements a globally-shared address space abstraction on top of a physically distributed memory substrate. In this paper, we discuss the development of a contention-aware network model intended to be integrated in a full-system XMT simulator. We start by measuring the effects of network contention in a 128-processor XMT machine and then investigate the trade-off that exists between simulation accuracy and speed, by comparing three network models which operate at different levels of accuracy. The comparison and model validation is performed by executing a string-matching algorithm on the full-system simulator and on the XMT, using three datasets that generate noticeably different contention patterns.« less

  12. Model-based object classification using unification grammars and abstract representations

    NASA Astrophysics Data System (ADS)

    Liburdy, Kathleen A.; Schalkoff, Robert J.

    1993-04-01

    The design and implementation of a high level computer vision system which performs object classification is described. General object labelling and functional analysis require models of classes which display a wide range of geometric variations. A large representational gap exists between abstract criteria such as `graspable' and current geometric image descriptions. The vision system developed and described in this work addresses this problem and implements solutions based on a fusion of semantics, unification, and formal language theory. Object models are represented using unification grammars, which provide a framework for the integration of structure and semantics. A methodology for the derivation of symbolic image descriptions capable of interacting with the grammar-based models is described and implemented. A unification-based parser developed for this system achieves object classification by determining if the symbolic image description can be unified with the abstract criteria of an object model. Future research directions are indicated.

  13. Modelling difficulties in abstract thinking in psychosis: the importance of socio-developmental background.

    PubMed

    Berg, A O; Melle, I; Zuber, V; Simonsen, C; Nerhus, M; Ueland, T; Andreassen, O A; Sundet, K; Vaskinn, A

    2017-01-01

    Abstract thinking is important in modern understanding of neurocognitive abilities, and a symptom of thought disorder in psychosis. In patients with psychosis, we assessed if socio-developmental background influences abstract thinking, and the association with executive functioning and clinical psychosis symptoms. Participants (n = 174) had a diagnosis of psychotic or bipolar disorder, were 17-65 years, intelligence quotient (IQ) > 70, fluent in a Scandinavian language, and their full primary education in Norway. Immigrants (N = 58) were matched (1:2) with participants without a history of migration (N = 116). All participants completed a neurocognitive and clinical assessment. Socio-developmental background was operationalised as human developmental index (HDI) of country of birth, at year of birth. Structural equation modelling was used to assess the model with best fit. The model with best fit, χ 2  = 96.591, df = 33, p < .001, confirmed a significant indirect effect of HDI scores on abstract thinking through executive functioning, but not through clinical psychosis symptoms. This study found that socio-developmental background influences abstract thinking in psychosis by indirect effect through executive functioning. We should take into account socio-developmental background in the interpretation of neurocognitive performance in patients with psychosis, and prioritise cognitive remediation in treatment of immigrant patients.

  14. Functional networks inference from rule-based machine learning models.

    PubMed

    Lazzarini, Nicola; Widera, Paweł; Williamson, Stuart; Heer, Rakesh; Krasnogor, Natalio; Bacardit, Jaume

    2016-01-01

    Functional networks play an important role in the analysis of biological processes and systems. The inference of these networks from high-throughput (-omics) data is an area of intense research. So far, the similarity-based inference paradigm (e.g. gene co-expression) has been the most popular approach. It assumes a functional relationship between genes which are expressed at similar levels across different samples. An alternative to this paradigm is the inference of relationships from the structure of machine learning models. These models are able to capture complex relationships between variables, that often are different/complementary to the similarity-based methods. We propose a protocol to infer functional networks from machine learning models, called FuNeL. It assumes, that genes used together within a rule-based machine learning model to classify the samples, might also be functionally related at a biological level. The protocol is first tested on synthetic datasets and then evaluated on a test suite of 8 real-world datasets related to human cancer. The networks inferred from the real-world data are compared against gene co-expression networks of equal size, generated with 3 different methods. The comparison is performed from two different points of view. We analyse the enriched biological terms in the set of network nodes and the relationships between known disease-associated genes in a context of the network topology. The comparison confirms both the biological relevance and the complementary character of the knowledge captured by the FuNeL networks in relation to similarity-based methods and demonstrates its potential to identify known disease associations as core elements of the network. Finally, using a prostate cancer dataset as a case study, we confirm that the biological knowledge captured by our method is relevant to the disease and consistent with the specialised literature and with an independent dataset not used in the inference process. The

  15. Kalman approach to accuracy management for interoperable heterogeneous model abstraction within an HLA-compliant simulation

    NASA Astrophysics Data System (ADS)

    Leskiw, Donald M.; Zhau, Junmei

    2000-06-01

    This paper reports on results from an ongoing project to develop methodologies for representing and managing multiple, concurrent levels of detail and enabling high performance computing using parallel arrays within distributed object-based simulation frameworks. At this time we present the methodology for representing and managing multiple, concurrent levels of detail and modeling accuracy by using a representation based on the Kalman approach for estimation. The Kalman System Model equations are used to represent model accuracy, Kalman Measurement Model equations provide transformations between heterogeneous levels of detail, and interoperability among disparate abstractions is provided using a form of the Kalman Update equations.

  16. Ontological modelling of knowledge management for human-machine integrated design of ultra-precision grinding machine

    NASA Astrophysics Data System (ADS)

    Hong, Haibo; Yin, Yuehong; Chen, Xing

    2016-11-01

    Despite the rapid development of computer science and information technology, an efficient human-machine integrated enterprise information system for designing complex mechatronic products is still not fully accomplished, partly because of the inharmonious communication among collaborators. Therefore, one challenge in human-machine integration is how to establish an appropriate knowledge management (KM) model to support integration and sharing of heterogeneous product knowledge. Aiming at the diversity of design knowledge, this article proposes an ontology-based model to reach an unambiguous and normative representation of knowledge. First, an ontology-based human-machine integrated design framework is described, then corresponding ontologies and sub-ontologies are established according to different purposes and scopes. Second, a similarity calculation-based ontology integration method composed of ontology mapping and ontology merging is introduced. The ontology searching-based knowledge sharing method is then developed. Finally, a case of human-machine integrated design of a large ultra-precision grinding machine is used to demonstrate the effectiveness of the method.

  17. Omnibus Risk Assessment via Accelerated Failure Time Kernel Machine Modeling

    PubMed Central

    Sinnott, Jennifer A.; Cai, Tianxi

    2013-01-01

    Summary Integrating genomic information with traditional clinical risk factors to improve the prediction of disease outcomes could profoundly change the practice of medicine. However, the large number of potential markers and possible complexity of the relationship between markers and disease make it difficult to construct accurate risk prediction models. Standard approaches for identifying important markers often rely on marginal associations or linearity assumptions and may not capture non-linear or interactive effects. In recent years, much work has been done to group genes into pathways and networks. Integrating such biological knowledge into statistical learning could potentially improve model interpretability and reliability. One effective approach is to employ a kernel machine (KM) framework, which can capture nonlinear effects if nonlinear kernels are used (Scholkopf and Smola, 2002; Liu et al., 2007, 2008). For survival outcomes, KM regression modeling and testing procedures have been derived under a proportional hazards (PH) assumption (Li and Luan, 2003; Cai et al., 2011). In this paper, we derive testing and prediction methods for KM regression under the accelerated failure time model, a useful alternative to the PH model. We approximate the null distribution of our test statistic using resampling procedures. When multiple kernels are of potential interest, it may be unclear in advance which kernel to use for testing and estimation. We propose a robust Omnibus Test that combines information across kernels, and an approach for selecting the best kernel for estimation. The methods are illustrated with an application in breast cancer. PMID:24328713

  18. Omnibus risk assessment via accelerated failure time kernel machine modeling.

    PubMed

    Sinnott, Jennifer A; Cai, Tianxi

    2013-12-01

    Integrating genomic information with traditional clinical risk factors to improve the prediction of disease outcomes could profoundly change the practice of medicine. However, the large number of potential markers and possible complexity of the relationship between markers and disease make it difficult to construct accurate risk prediction models. Standard approaches for identifying important markers often rely on marginal associations or linearity assumptions and may not capture non-linear or interactive effects. In recent years, much work has been done to group genes into pathways and networks. Integrating such biological knowledge into statistical learning could potentially improve model interpretability and reliability. One effective approach is to employ a kernel machine (KM) framework, which can capture nonlinear effects if nonlinear kernels are used (Scholkopf and Smola, 2002; Liu et al., 2007, 2008). For survival outcomes, KM regression modeling and testing procedures have been derived under a proportional hazards (PH) assumption (Li and Luan, 2003; Cai, Tonini, and Lin, 2011). In this article, we derive testing and prediction methods for KM regression under the accelerated failure time (AFT) model, a useful alternative to the PH model. We approximate the null distribution of our test statistic using resampling procedures. When multiple kernels are of potential interest, it may be unclear in advance which kernel to use for testing and estimation. We propose a robust Omnibus Test that combines information across kernels, and an approach for selecting the best kernel for estimation. The methods are illustrated with an application in breast cancer. © 2013, The International Biometric Society.

  19. Risk Classification with an Adaptive Naive Bayes Kernel Machine Model.

    PubMed

    Minnier, Jessica; Yuan, Ming; Liu, Jun S; Cai, Tianxi

    2015-04-22

    Genetic studies of complex traits have uncovered only a small number of risk markers explaining a small fraction of heritability and adding little improvement to disease risk prediction. Standard single marker methods may lack power in selecting informative markers or estimating effects. Most existing methods also typically do not account for non-linearity. Identifying markers with weak signals and estimating their joint effects among many non-informative markers remains challenging. One potential approach is to group markers based on biological knowledge such as gene structure. If markers in a group tend to have similar effects, proper usage of the group structure could improve power and efficiency in estimation. We propose a two-stage method relating markers to disease risk by taking advantage of known gene-set structures. Imposing a naive bayes kernel machine (KM) model, we estimate gene-set specific risk models that relate each gene-set to the outcome in stage I. The KM framework efficiently models potentially non-linear effects of predictors without requiring explicit specification of functional forms. In stage II, we aggregate information across gene-sets via a regularization procedure. Estimation and computational efficiency is further improved with kernel principle component analysis. Asymptotic results for model estimation and gene set selection are derived and numerical studies suggest that the proposed procedure could outperform existing procedures for constructing genetic risk models.

  20. Modeling of tool path for the CNC sheet cutting machines

    NASA Astrophysics Data System (ADS)

    Petunin, Aleksandr A.

    2015-11-01

    In the paper the problem of tool path optimization for CNC (Computer Numerical Control) cutting machines is considered. The classification of the cutting techniques is offered. We also propose a new classification of toll path problems. The tasks of cost minimization and time minimization for standard cutting technique (Continuous Cutting Problem, CCP) and for one of non-standard cutting techniques (Segment Continuous Cutting Problem, SCCP) are formalized. We show that the optimization tasks can be interpreted as discrete optimization problem (generalized travel salesman problem with additional constraints, GTSP). Formalization of some constraints for these tasks is described. For the solution GTSP we offer to use mathematical model of Prof. Chentsov based on concept of a megalopolis and dynamic programming.

  1. Programming with models: modularity and abstraction provide powerful capabilities for systems biology

    PubMed Central

    Mallavarapu, Aneil; Thomson, Matthew; Ullian, Benjamin; Gunawardena, Jeremy

    2008-01-01

    Mathematical models are increasingly used to understand how phenotypes emerge from systems of molecular interactions. However, their current construction as monolithic sets of equations presents a fundamental barrier to progress. Overcoming this requires modularity, enabling sub-systems to be specified independently and combined incrementally, and abstraction, enabling generic properties of biological processes to be specified independently of specific instances. These, in turn, require models to be represented as programs rather than as datatypes. Programmable modularity and abstraction enables libraries of modules to be created, which can be instantiated and reused repeatedly in different contexts with different components. We have developed a computational infrastructure that accomplishes this. We show here why such capabilities are needed, what is required to implement them and what can be accomplished with them that could not be done previously. PMID:18647734

  2. Programming with models: modularity and abstraction provide powerful capabilities for systems biology.

    PubMed

    Mallavarapu, Aneil; Thomson, Matthew; Ullian, Benjamin; Gunawardena, Jeremy

    2009-03-06

    Mathematical models are increasingly used to understand how phenotypes emerge from systems of molecular interactions. However, their current construction as monolithic sets of equations presents a fundamental barrier to progress. Overcoming this requires modularity, enabling sub-systems to be specified independently and combined incrementally, and abstraction, enabling generic properties of biological processes to be specified independently of specific instances. These, in turn, require models to be represented as programs rather than as datatypes. Programmable modularity and abstraction enables libraries of modules to be created, which can be instantiated and reused repeatedly in different contexts with different components. We have developed a computational infrastructure that accomplishes this. We show here why such capabilities are needed, what is required to implement them and what can be accomplished with them that could not be done previously.

  3. A minimal kinetic model for a viral DNA packaging machine.

    PubMed

    Yang, Qin; Catalano, Carlos Enrique

    2004-01-20

    Terminase enzymes are common to both eukaryotic and prokaryotic double-stranded DNA viruses. These enzymes possess ATPase and nuclease activities that work in concert to "package" a viral genome into an empty procapsid, and it is likely that terminase enzymes from disparate viruses utilize a common packaging mechanism. Bacteriophage lambda terminase possesses a site-specific nuclease activity, a so-called helicase activity, a DNA translocase activity, and multiple ATPase catalytic sites that function to package viral DNA. Allosteric interactions between the multiple catalytic sites have been reported. This study probes these catalytic interactions using enzyme kinetic, photoaffinity labeling, and vanadate inhibition studies. The ensemble of data forms the basis for a minimal kinetic model for lambda terminase. The model incorporates an ADP-driven conformational reorganization of the terminase subunits assembled on viral DNA, which is central to the activation of a catalytically competent packaging machine. The proposed model provides a unifying mechanism for allosteric interaction between the multiple catalytic sites of the holoenzyme and explains much of the kinetic data in the literature. Given that similar packaging mechanisms have been proposed for viruses as dissimilar as lambda and the herpes viruses, the model may find general utility in our global understanding of the enzymology of virus assembly.

  4. Kinetic modeling of α-hydrogen abstractions from unsaturated and saturated oxygenate compounds by hydrogen atoms.

    PubMed

    Paraskevas, Paschalis D; Sabbe, Maarten K; Reyniers, Marie-Françoise; Papayannakos, Nikos G; Marin, Guy B

    2014-10-09

    Hydrogen-abstraction reactions play a significant role in thermal biomass conversion processes, as well as regular gasification, pyrolysis, or combustion. In this work, a group additivity model is constructed that allows prediction of reaction rates and Arrhenius parameters of hydrogen abstractions by hydrogen atoms from alcohols, ethers, esters, peroxides, ketones, aldehydes, acids, and diketones in a broad temperature range (300-2000 K). A training set of 60 reactions was developed with rate coefficients and Arrhenius parameters calculated by the CBS-QB3 method in the high-pressure limit with tunneling corrections using Eckart tunneling coefficients. From this set of reactions, 15 group additive values were derived for the forward and the reverse reaction, 4 referring to primary and 11 to secondary contributions. The accuracy of the model is validated upon an ab initio and an experimental validation set of 19 and 21 reaction rates, respectively, showing that reaction rates can be predicted with a mean factor of deviation of 2 for the ab initio and 3 for the experimental values. Hence, this work illustrates that the developed group additive model can be reliably applied for the accurate prediction of kinetics of α-hydrogen abstractions by hydrogen atoms from a broad range of oxygenates.

  5. An initial-abstraction, constant-loss model for unit hydrograph modeling for applicable watersheds in Texas

    USGS Publications Warehouse

    Asquith, William H.; Roussel, Meghan C.

    2007-01-01

    Estimation of representative hydrographs from design storms, which are known as design hydrographs, provides for cost-effective, riskmitigated design of drainage structures such as bridges, culverts, roadways, and other infrastructure. During 2001?07, the U.S. Geological Survey (USGS), in cooperation with the Texas Department of Transportation, investigated runoff hydrographs, design storms, unit hydrographs,and watershed-loss models to enhance design hydrograph estimation in Texas. Design hydrographs ideally should mimic the general volume, peak, and shape of observed runoff hydrographs. Design hydrographs commonly are estimated in part by unit hydrographs. A unit hydrograph is defined as the runoff hydrograph that results from a unit pulse of excess rainfall uniformly distributed over the watershed at a constant rate for a specific duration. A time-distributed, watershed-loss model is required for modeling by unit hydrographs. This report develops a specific time-distributed, watershed-loss model known as an initial-abstraction, constant-loss model. For this watershed-loss model, a watershed is conceptualized to have the capacity to store or abstract an absolute depth of rainfall at and near the beginning of a storm. Depths of total rainfall less than this initial abstraction do not produce runoff. The watershed also is conceptualized to have the capacity to remove rainfall at a constant rate (loss) after the initial abstraction is satisfied. Additional rainfall inputs after the initial abstraction is satisfied contribute to runoff if the rainfall rate (intensity) is larger than the constant loss. The initial abstraction, constant-loss model thus is a two-parameter model. The initial-abstraction, constant-loss model is investigated through detailed computational and statistical analysis of observed rainfall and runoff data for 92 USGS streamflow-gaging stations (watersheds) in Texas with contributing drainage areas from 0.26 to 166 square miles. The analysis is

  6. Machine Learning Methods for Analysis of Metabolic Data and Metabolic Pathway Modeling

    PubMed Central

    Cuperlovic-Culf, Miroslava

    2018-01-01

    Machine learning uses experimental data to optimize clustering or classification of samples or features, or to develop, augment or verify models that can be used to predict behavior or properties of systems. It is expected that machine learning will help provide actionable knowledge from a variety of big data including metabolomics data, as well as results of metabolism models. A variety of machine learning methods has been applied in bioinformatics and metabolism analyses including self-organizing maps, support vector machines, the kernel machine, Bayesian networks or fuzzy logic. To a lesser extent, machine learning has also been utilized to take advantage of the increasing availability of genomics and metabolomics data for the optimization of metabolic network models and their analysis. In this context, machine learning has aided the development of metabolic networks, the calculation of parameters for stoichiometric and kinetic models, as well as the analysis of major features in the model for the optimal application of bioreactors. Examples of this very interesting, albeit highly complex, application of machine learning for metabolism modeling will be the primary focus of this review presenting several different types of applications for model optimization, parameter determination or system analysis using models, as well as the utilization of several different types of machine learning technologies. PMID:29324649

  7. Machine Learning Methods for Analysis of Metabolic Data and Metabolic Pathway Modeling.

    PubMed

    Cuperlovic-Culf, Miroslava

    2018-01-11

    Machine learning uses experimental data to optimize clustering or classification of samples or features, or to develop, augment or verify models that can be used to predict behavior or properties of systems. It is expected that machine learning will help provide actionable knowledge from a variety of big data including metabolomics data, as well as results of metabolism models. A variety of machine learning methods has been applied in bioinformatics and metabolism analyses including self-organizing maps, support vector machines, the kernel machine, Bayesian networks or fuzzy logic. To a lesser extent, machine learning has also been utilized to take advantage of the increasing availability of genomics and metabolomics data for the optimization of metabolic network models and their analysis. In this context, machine learning has aided the development of metabolic networks, the calculation of parameters for stoichiometric and kinetic models, as well as the analysis of major features in the model for the optimal application of bioreactors. Examples of this very interesting, albeit highly complex, application of machine learning for metabolism modeling will be the primary focus of this review presenting several different types of applications for model optimization, parameter determination or system analysis using models, as well as the utilization of several different types of machine learning technologies.

  8. Modeling the Virtual Machine Launching Overhead under Fermicloud

    SciTech Connect

    Garzoglio, Gabriele; Wu, Hao; Ren, Shangping

    FermiCloud is a private cloud developed by the Fermi National Accelerator Laboratory for scientific workflows. The Cloud Bursting module of the FermiCloud enables the FermiCloud, when more computational resources are needed, to automatically launch virtual machines to available resources such as public clouds. One of the main challenges in developing the cloud bursting module is to decide when and where to launch a VM so that all resources are most effectively and efficiently utilized and the system performance is optimized. However, based on FermiCloud’s system operational data, the VM launching overhead is not a constant. It varies with physical resourcemore » (CPU, memory, I/O device) utilization at the time when a VM is launched. Hence, to make judicious decisions as to when and where a VM should be launched, a VM launch overhead reference model is needed. The paper is to develop a VM launch overhead reference model based on operational data we have obtained on FermiCloud and uses the reference model to guide the cloud bursting process.« less

  9. Derivative Free Optimization of Complex Systems with the Use of Statistical Machine Learning Models

    DTIC Science & Technology

    2015-09-12

    AFRL-AFOSR-VA-TR-2015-0278 DERIVATIVE FREE OPTIMIZATION OF COMPLEX SYSTEMS WITH THE USE OF STATISTICAL MACHINE LEARNING MODELS Katya Scheinberg...COMPLEX SYSTEMS WITH THE USE OF STATISTICAL MACHINE LEARNING MODELS 5a.  CONTRACT NUMBER 5b.  GRANT NUMBER FA9550-11-1-0239 5c.  PROGRAM ELEMENT...developed, which has been the focus of our research. 15. SUBJECT TERMS optimization, Derivative-Free Optimization, Statistical Machine Learning 16. SECURITY

  10. Abstract probabilistic CNOT gate model based on double encoding: study of the errors and physical realizability

    NASA Astrophysics Data System (ADS)

    Gueddana, Amor; Attia, Moez; Chatta, Rihab

    2015-03-01

    In this work, we study the error sources standing behind the non-perfect linear optical quantum components composing a non-deterministic quantum CNOT gate model, which performs the CNOT function with a success probability of 4/27 and uses a double encoding technique to represent photonic qubits at the control and the target. We generalize this model to an abstract probabilistic CNOT version and determine the realizability limits depending on a realistic range of the errors. Finally, we discuss physical constraints allowing the implementation of the Asymmetric Partially Polarizing Beam Splitter (APPBS), which is at the heart of correctly realizing the CNOT function.

  11. Parameterizing Phrase Based Statistical Machine Translation Models: An Analytic Study

    ERIC Educational Resources Information Center

    Cer, Daniel

    2011-01-01

    The goal of this dissertation is to determine the best way to train a statistical machine translation system. I first develop a state-of-the-art machine translation system called Phrasal and then use it to examine a wide variety of potential learning algorithms and optimization criteria and arrive at two very surprising results. First, despite the…

  12. Access, Equity, and Opportunity. Women in Machining: A Model Program.

    ERIC Educational Resources Information Center

    Warner, Heather

    The Women in Machining (WIM) program is a Machine Action Project (MAP) initiative that was developed in response to a local skilled metalworking labor shortage, despite a virtual absence of women and people of color from area shops. The project identified post-war stereotypes and other barriers that must be addressed if women are to have an equal…

  13. Modeling and simulation of five-axis virtual machine based on NX

    NASA Astrophysics Data System (ADS)

    Li, Xiaoda; Zhan, Xianghui

    2018-04-01

    Virtual technology in the machinery manufacturing industry has shown the role of growing. In this paper, the Siemens NX software is used to model the virtual CNC machine tool, and the parameters of the virtual machine are defined according to the actual parameters of the machine tool so that the virtual simulation can be carried out without loss of the accuracy of the simulation. How to use the machine builder of the CAM module to define the kinematic chain and machine components of the machine is described. The simulation of virtual machine can provide alarm information of tool collision and over cutting during the process to users, and can evaluate and forecast the rationality of the technological process.

  14. Predictive Surface Roughness Model for End Milling of Machinable Glass Ceramic

    NASA Astrophysics Data System (ADS)

    Mohan Reddy, M.; Gorin, Alexander; Abou-El-Hossein, K. A.

    2011-02-01

    Advanced ceramics of Machinable glass ceramic is attractive material to produce high accuracy miniaturized components for many applications in various industries such as aerospace, electronics, biomedical, automotive and environmental communications due to their wear resistance, high hardness, high compressive strength, good corrosion resistance and excellent high temperature properties. Many research works have been conducted in the last few years to investigate the performance of different machining operations when processing various advanced ceramics. Micro end-milling is one of the machining methods to meet the demand of micro parts. Selecting proper machining parameters are important to obtain good surface finish during machining of Machinable glass ceramic. Therefore, this paper describes the development of predictive model for the surface roughness of Machinable glass ceramic in terms of speed, feed rate by using micro end-milling operation.

  15. Modeling stochastic kinetics of molecular machines at multiple levels: from molecules to modules.

    PubMed

    Chowdhury, Debashish

    2013-06-04

    A molecular machine is either a single macromolecule or a macromolecular complex. In spite of the striking superficial similarities between these natural nanomachines and their man-made macroscopic counterparts, there are crucial differences. Molecular machines in a living cell operate stochastically in an isothermal environment far from thermodynamic equilibrium. In this mini-review we present a catalog of the molecular machines and an inventory of the essential toolbox for theoretically modeling these machines. The tool kits include 1), nonequilibrium statistical-physics techniques for modeling machines and machine-driven processes; and 2), statistical-inference methods for reverse engineering a functional machine from the empirical data. The cell is often likened to a microfactory in which the machineries are organized in modular fashion; each module consists of strongly coupled multiple machines, but different modules interact weakly with each other. This microfactory has its own automated supply chain and delivery system. Buoyed by the success achieved in modeling individual molecular machines, we advocate integration of these models in the near future to develop models of functional modules. A system-level description of the cell from the perspective of molecular machinery (the mechanome) is likely to emerge from further integrations that we envisage here. Copyright © 2013 Biophysical Society. Published by Elsevier Inc. All rights reserved.

  16. Modeling Stochastic Kinetics of Molecular Machines at Multiple Levels: From Molecules to Modules

    PubMed Central

    Chowdhury, Debashish

    2013-01-01

    A molecular machine is either a single macromolecule or a macromolecular complex. In spite of the striking superficial similarities between these natural nanomachines and their man-made macroscopic counterparts, there are crucial differences. Molecular machines in a living cell operate stochastically in an isothermal environment far from thermodynamic equilibrium. In this mini-review we present a catalog of the molecular machines and an inventory of the essential toolbox for theoretically modeling these machines. The tool kits include 1), nonequilibrium statistical-physics techniques for modeling machines and machine-driven processes; and 2), statistical-inference methods for reverse engineering a functional machine from the empirical data. The cell is often likened to a microfactory in which the machineries are organized in modular fashion; each module consists of strongly coupled multiple machines, but different modules interact weakly with each other. This microfactory has its own automated supply chain and delivery system. Buoyed by the success achieved in modeling individual molecular machines, we advocate integration of these models in the near future to develop models of functional modules. A system-level description of the cell from the perspective of molecular machinery (the mechanome) is likely to emerge from further integrations that we envisage here. PMID:23746505

  17. Generating Models of Infinite-State Communication Protocols Using Regular Inference with Abstraction

    NASA Astrophysics Data System (ADS)

    Aarts, Fides; Jonsson, Bengt; Uijen, Johan

    In order to facilitate model-based verification and validation, effort is underway to develop techniques for generating models of communication system components from observations of their external behavior. Most previous such work has employed regular inference techniques which generate modest-size finite-state models. They typically suppress parameters of messages, although these have a significant impact on control flow in many communication protocols. We present a framework, which adapts regular inference to include data parameters in messages and states for generating components with large or infinite message alphabets. A main idea is to adapt the framework of predicate abstraction, successfully used in formal verification. Since we are in a black-box setting, the abstraction must be supplied externally, using information about how the component manages data parameters. We have implemented our techniques by connecting the LearnLib tool for regular inference with the protocol simulator ns-2, and generated a model of the SIP component as implemented in ns-2.

  18. Data Visualization Saliency Model: A Tool for Evaluating Abstract Data Visualizations

    DOE PAGES

    Matzen, Laura E.; Haass, Michael J.; Divis, Kristin M.; ...

    2017-08-29

    Evaluating the effectiveness of data visualizations is a challenging undertaking and often relies on one-off studies that test a visualization in the context of one specific task. Researchers across the fields of data science, visualization, and human-computer interaction are calling for foundational tools and principles that could be applied to assessing the effectiveness of data visualizations in a more rapid and generalizable manner. One possibility for such a tool is a model of visual saliency for data visualizations. Visual saliency models are typically based on the properties of the human visual cortex and predict which areas of a scene havemore » visual features (e.g. color, luminance, edges) that are likely to draw a viewer's attention. While these models can accurately predict where viewers will look in a natural scene, they typically do not perform well for abstract data visualizations. In this paper, we discuss the reasons for the poor performance of existing saliency models when applied to data visualizations. We introduce the Data Visualization Saliency (DVS) model, a saliency model tailored to address some of these weaknesses, and we test the performance of the DVS model and existing saliency models by comparing the saliency maps produced by the models to eye tracking data obtained from human viewers. In conclusion, we describe how modified saliency models could be used as general tools for assessing the effectiveness of visualizations, including the strengths and weaknesses of this approach.« less

  19. Data Visualization Saliency Model: A Tool for Evaluating Abstract Data Visualizations

    SciTech Connect

    Matzen, Laura E.; Haass, Michael J.; Divis, Kristin M.

    Evaluating the effectiveness of data visualizations is a challenging undertaking and often relies on one-off studies that test a visualization in the context of one specific task. Researchers across the fields of data science, visualization, and human-computer interaction are calling for foundational tools and principles that could be applied to assessing the effectiveness of data visualizations in a more rapid and generalizable manner. One possibility for such a tool is a model of visual saliency for data visualizations. Visual saliency models are typically based on the properties of the human visual cortex and predict which areas of a scene havemore » visual features (e.g. color, luminance, edges) that are likely to draw a viewer's attention. While these models can accurately predict where viewers will look in a natural scene, they typically do not perform well for abstract data visualizations. In this paper, we discuss the reasons for the poor performance of existing saliency models when applied to data visualizations. We introduce the Data Visualization Saliency (DVS) model, a saliency model tailored to address some of these weaknesses, and we test the performance of the DVS model and existing saliency models by comparing the saliency maps produced by the models to eye tracking data obtained from human viewers. In conclusion, we describe how modified saliency models could be used as general tools for assessing the effectiveness of visualizations, including the strengths and weaknesses of this approach.« less

  20. Alternative Models of Service, Centralized Machine Operations. Phase II Report. Volume II.

    ERIC Educational Resources Information Center

    Technology Management Corp., Alexandria, VA.

    A study was conducted to determine if the centralization of playback machine operations for the national free library program would be feasible, economical, and desirable. An alternative model of playback machine services was constructed and compared with existing network operations considering both cost and service. The alternative model was…

  1. AZOrange - High performance open source machine learning for QSAR modeling in a graphical programming environment

    PubMed Central

    2011-01-01

    Background Machine learning has a vast range of applications. In particular, advanced machine learning methods are routinely and increasingly used in quantitative structure activity relationship (QSAR) modeling. QSAR data sets often encompass tens of thousands of compounds and the size of proprietary, as well as public data sets, is rapidly growing. Hence, there is a demand for computationally efficient machine learning algorithms, easily available to researchers without extensive machine learning knowledge. In granting the scientific principles of transparency and reproducibility, Open Source solutions are increasingly acknowledged by regulatory authorities. Thus, an Open Source state-of-the-art high performance machine learning platform, interfacing multiple, customized machine learning algorithms for both graphical programming and scripting, to be used for large scale development of QSAR models of regulatory quality, is of great value to the QSAR community. Results This paper describes the implementation of the Open Source machine learning package AZOrange. AZOrange is specially developed to support batch generation of QSAR models in providing the full work flow of QSAR modeling, from descriptor calculation to automated model building, validation and selection. The automated work flow relies upon the customization of the machine learning algorithms and a generalized, automated model hyper-parameter selection process. Several high performance machine learning algorithms are interfaced for efficient data set specific selection of the statistical method, promoting model accuracy. Using the high performance machine learning algorithms of AZOrange does not require programming knowledge as flexible applications can be created, not only at a scripting level, but also in a graphical programming environment. Conclusions AZOrange is a step towards meeting the needs for an Open Source high performance machine learning platform, supporting the efficient development of

  2. AZOrange - High performance open source machine learning for QSAR modeling in a graphical programming environment.

    PubMed

    Stålring, Jonna C; Carlsson, Lars A; Almeida, Pedro; Boyer, Scott

    2011-07-28

    Machine learning has a vast range of applications. In particular, advanced machine learning methods are routinely and increasingly used in quantitative structure activity relationship (QSAR) modeling. QSAR data sets often encompass tens of thousands of compounds and the size of proprietary, as well as public data sets, is rapidly growing. Hence, there is a demand for computationally efficient machine learning algorithms, easily available to researchers without extensive machine learning knowledge. In granting the scientific principles of transparency and reproducibility, Open Source solutions are increasingly acknowledged by regulatory authorities. Thus, an Open Source state-of-the-art high performance machine learning platform, interfacing multiple, customized machine learning algorithms for both graphical programming and scripting, to be used for large scale development of QSAR models of regulatory quality, is of great value to the QSAR community. This paper describes the implementation of the Open Source machine learning package AZOrange. AZOrange is specially developed to support batch generation of QSAR models in providing the full work flow of QSAR modeling, from descriptor calculation to automated model building, validation and selection. The automated work flow relies upon the customization of the machine learning algorithms and a generalized, automated model hyper-parameter selection process. Several high performance machine learning algorithms are interfaced for efficient data set specific selection of the statistical method, promoting model accuracy. Using the high performance machine learning algorithms of AZOrange does not require programming knowledge as flexible applications can be created, not only at a scripting level, but also in a graphical programming environment. AZOrange is a step towards meeting the needs for an Open Source high performance machine learning platform, supporting the efficient development of highly accurate QSAR models

  3. A machine learning model with human cognitive biases capable of learning from small and biased datasets.

    PubMed

    Taniguchi, Hidetaka; Sato, Hiroshi; Shirakawa, Tomohiro

    2018-05-09

    Human learners can generalize a new concept from a small number of samples. In contrast, conventional machine learning methods require large amounts of data to address the same types of problems. Humans have cognitive biases that promote fast learning. Here, we developed a method to reduce the gap between human beings and machines in this type of inference by utilizing cognitive biases. We implemented a human cognitive model into machine learning algorithms and compared their performance with the currently most popular methods, naïve Bayes, support vector machine, neural networks, logistic regression and random forests. We focused on the task of spam classification, which has been studied for a long time in the field of machine learning and often requires a large amount of data to obtain high accuracy. Our models achieved superior performance with small and biased samples in comparison with other representative machine learning methods.

  4. Multi-objective optimization model of CNC machining to minimize processing time and environmental impact

    NASA Astrophysics Data System (ADS)

    Hamada, Aulia; Rosyidi, Cucuk Nur; Jauhari, Wakhid Ahmad

    2017-11-01

    Minimizing processing time in a production system can increase the efficiency of a manufacturing company. Processing time are influenced by application of modern technology and machining parameter. Application of modern technology can be apply by use of CNC machining, one of the machining process can be done with a CNC machining is turning. However, the machining parameters not only affect the processing time but also affect the environmental impact. Hence, optimization model is needed to optimize the machining parameters to minimize the processing time and environmental impact. This research developed a multi-objective optimization to minimize the processing time and environmental impact in CNC turning process which will result in optimal decision variables of cutting speed and feed rate. Environmental impact is converted from environmental burden through the use of eco-indicator 99. The model were solved by using OptQuest optimization software from Oracle Crystal Ball.

  5. Support vector machines-based modelling of seismic liquefaction potential

    NASA Astrophysics Data System (ADS)

    Pal, Mahesh

    2006-08-01

    This paper investigates the potential of support vector machines (SVM)-based classification approach to assess the liquefaction potential from actual standard penetration test (SPT) and cone penetration test (CPT) field data. SVMs are based on statistical learning theory and found to work well in comparison to neural networks in several other applications. Both CPT and SPT field data sets is used with SVMs for predicting the occurrence and non-occurrence of liquefaction based on different input parameter combination. With SPT and CPT test data sets, highest accuracy of 96 and 97%, respectively, was achieved with SVMs. This suggests that SVMs can effectively be used to model the complex relationship between different soil parameter and the liquefaction potential. Several other combinations of input variable were used to assess the influence of different input parameters on liquefaction potential. Proposed approach suggest that neither normalized cone resistance value with CPT data nor the calculation of standardized SPT value is required with SPT data. Further, SVMs required few user-defined parameters and provide better performance in comparison to neural network approach.

  6. Developing robust arsenic awareness prediction models using machine learning algorithms.

    PubMed

    Singh, Sushant K; Taylor, Robert W; Rahman, Mohammad Mahmudur; Pradhan, Biswajeet

    2018-04-01

    Arsenic awareness plays a vital role in ensuring the sustainability of arsenic mitigation technologies. Thus far, however, few studies have dealt with the sustainability of such technologies and its associated socioeconomic dimensions. As a result, arsenic awareness prediction has not yet been fully conceptualized. Accordingly, this study evaluated arsenic awareness among arsenic-affected communities in rural India, using a structured questionnaire to record socioeconomic, demographic, and other sociobehavioral factors with an eye to assessing their association with and influence on arsenic awareness. First a logistic regression model was applied and its results compared with those produced by six state-of-the-art machine-learning algorithms (Support Vector Machine [SVM], Kernel-SVM, Decision Tree [DT], k-Nearest Neighbor [k-NN], Naïve Bayes [NB], and Random Forests [RF]) as measured by their accuracy at predicting arsenic awareness. Most (63%) of the surveyed population was found to be arsenic-aware. Significant arsenic awareness predictors were divided into three types: (1) socioeconomic factors: caste, education level, and occupation; (2) water and sanitation behavior factors: number of family members involved in water collection, distance traveled and time spent for water collection, places for defecation, and materials used for handwashing after defecation; and (3) social capital and trust factors: presence of anganwadi and people's trust in other community members, NGOs, and private agencies. Moreover, individuals' having higher social network positively contributed to arsenic awareness in the communities. Results indicated that both the SVM and the RF algorithms outperformed at overall prediction of arsenic awareness-a nonlinear classification problem. Lower-caste, less educated, and unemployed members of the population were found to be the most vulnerable, requiring immediate arsenic mitigation. To this end, local social institutions and NGOs could play a

  7. Empirical Models for Quantification of Machining Damage in Composite Materials

    NASA Astrophysics Data System (ADS)

    Machado, Carla Maria Moreira

    The tremendous growth which occurs at a global level of demand and use of composite materials brings with the need to develop new manufacturing tools and methodologies. One of the major uses of such materials, in particular plastics reinforced with carbon fibres, is their application in structural components for the aircraft industry with low weight and high stiffness. These components are produced in near-final form but the so-called secondary processes such as machining are often unavoidable. In this type of industry, drilling is the most frequent operation due to the need to obtain holes for riveting and fastening bolt assembly of structures. However, the problems arising from drilling, particularly the damage caused during the operation, may lead to rejection of components because it is an origin of lack of resistance. The delamination is the most important damage, as it causes a decrease of the mechanical properties of the components of an assembly and, irrefutably, a reduction of its reliability in use. It can also raise problems with regard to the tolerances of the assemblies. Moreover, the high speed machining is increasingly recognized to be a manufacturing technology that promotes productivity by reducing production times. However, the investigation whose focus is in high speed drilling is quite limited, and few studies on this subject have been found in the literature review. Thus, this thesis aims to investigate the effects of process variables in high speed drilling on the damage produced. The empirical models that relate the delamination damage, the thrust force and the torque with the process parameters were established using Response Surface Methodology. The process parameters considered as input factors were the spindle speed, the feed per tooth, the tool diameter and the workpiece thickness. A new method for fixing the workpiece was developed and tested. The results proved to be very promising since in the same cutting conditions and with this new

  8. Modelling the sensitivity of river reaches to water abstraction: RAPHSA- a hydroecology tool for environmental managers

    NASA Astrophysics Data System (ADS)

    Klaar, Megan; Laize, Cedric; Maddock, Ian; Acreman, Mike; Tanner, Kath; Peet, Sarah

    2014-05-01

    A key challenge for environmental managers is the determination of environmental flows which allow a maximum yield of water resources to be taken from surface and sub-surface sources, whilst ensuring sufficient water remains in the environment to support biota and habitats. It has long been known that sensitivity to changes in water levels resulting from river and groundwater abstractions varies between rivers. Whilst assessment at the catchment scale is ideal for determining broad pressures on water resources and ecosystems, assessment of the sensitivity of reaches to changes in flow has previously been done on a site-by-site basis, often with the application of detailed but time consuming techniques (e.g. PHABSIM). While this is appropriate for a limited number of sites, it is costly in terms of money and time resources and therefore not appropriate for application at a national level required by responsible licensing authorities. To address this need, the Environment Agency (England) is developing an operational tool to predict relationships between physical habitat and flow which may be applied by field staff to rapidly determine the sensitivity of physical habitat to flow alteration for use in water resource management planning. An initial model of river sensitivity to abstraction (defined as the change in physical habitat related to changes in river discharge) was developed using site characteristics and data from 66 individual PHABSIM surveys throughout the UK (Booker & Acreman, 2008). By applying a multivariate multiple linear regression analysis to the data to define habitat availability-flow curves using resource intensity as predictor variables, the model (known as RAPHSA- Rapid Assessment of Physical Habitat Sensitivity to Abstraction) is able to take a risk-based approach to modeled certainty. Site specific information gathered using desk-based, or a variable amount of field work can be used to predict the shape of the habitat- flow curves, with the

  9. DFT modeling of chemistry on the Z machine

    NASA Astrophysics Data System (ADS)

    Mattsson, Thomas

    2013-06-01

    Density Functional Theory (DFT) has proven remarkably accurate in predicting properties of matter under shock compression for a wide-range of elements and compounds: from hydrogen to xenon via water. Materials where chemistry plays a role are of particular interest for many applications. For example the deep interiors of Neptune, Uranus, and hundreds of similar exoplanets are composed of molecular ices of carbon, hydrogen, oxygen, and nitrogen at pressures of several hundred GPa and temperatures of many thousand Kelvin. High-quality thermophysical experimental data and high-fidelity simulations including chemical reaction are necessary to constrain planetary models over a large range of conditions. As examples of where chemical reactions are important, and demonstration of the high fidelity possible for these both structurally and chemically complex systems, we will discuss shock- and re-shock of liquid carbon dioxide (CO2) in the range 100 to 800 GPa, shock compression of the hydrocarbon polymers polyethylene (PE) and poly(4-methyl-1-pentene) (PMP), and finally simulations of shock compression of glow discharge polymer (GDP) including the effects of doping with germanium. Experimental results from Sandia's Z machine have time and again validated the DFT simulations at extreme conditions and the combination of experiment and DFT provide reliable data for evaluating existing and constructing future wide-range equations of state models for molecular compounds like CO2 and polymers like PE, PMP, and GDP. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Company, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.

  10. Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary View

    PubMed Central

    2016-01-01

    Background As more and more researchers are turning to big data for new opportunities of biomedical discoveries, machine learning models, as the backbone of big data analysis, are mentioned more often in biomedical journals. However, owing to the inherent complexity of machine learning methods, they are prone to misuse. Because of the flexibility in specifying machine learning models, the results are often insufficiently reported in research articles, hindering reliable assessment of model validity and consistent interpretation of model outputs. Objective To attain a set of guidelines on the use of machine learning predictive models within clinical settings to make sure the models are correctly applied and sufficiently reported so that true discoveries can be distinguished from random coincidence. Methods A multidisciplinary panel of machine learning experts, clinicians, and traditional statisticians were interviewed, using an iterative process in accordance with the Delphi method. Results The process produced a set of guidelines that consists of (1) a list of reporting items to be included in a research article and (2) a set of practical sequential steps for developing predictive models. Conclusions A set of guidelines was generated to enable correct application of machine learning models and consistent reporting of model specifications and results in biomedical research. We believe that such guidelines will accelerate the adoption of big data analysis, particularly with machine learning methods, in the biomedical research community. PMID:27986644

  11. Machine learning in updating predictive models of planning and scheduling transportation projects

    DOT National Transportation Integrated Search

    1997-01-01

    A method combining machine learning and regression analysis to automatically and intelligently update predictive models used in the Kansas Department of Transportations (KDOTs) internal management system is presented. The predictive models used...

  12. Development of an Empirical Model for Optimization of Machining Parameters to Minimize Power Consumption

    NASA Astrophysics Data System (ADS)

    Kant Garg, Girish; Garg, Suman; Sangwan, K. S.

    2018-04-01

    The manufacturing sector consumes huge energy demand and the machine tools used in this sector have very less energy efficiency. Selection of the optimum machining parameters for machine tools is significant for energy saving and for reduction of environmental emission. In this work an empirical model is developed to minimize the power consumption using response surface methodology. The experiments are performed on a lathe machine tool during the turning of AISI 6061 Aluminum with coated tungsten inserts. The relationship between the power consumption and machining parameters is adequately modeled. This model is used for formulation of minimum power consumption criterion as a function of optimal machining parameters using desirability function approach. The influence of machining parameters on the energy consumption has been found using the analysis of variance. The validation of the developed empirical model is proved using the confirmation experiments. The results indicate that the developed model is effective and has potential to be adopted by the industry for minimum power consumption of machine tools.

  13. Job shop scheduling model for non-identic machine with fixed delivery time to minimize tardiness

    NASA Astrophysics Data System (ADS)

    Kusuma, K. K.; Maruf, A.

    2016-02-01

    Scheduling non-identic machines problem with low utilization characteristic and fixed delivery time are frequent in manufacture industry. This paper propose a mathematical model to minimize total tardiness for non-identic machines in job shop environment. This model will be categorized as an integer linier programming model and using branch and bound algorithm as the solver method. We will use fixed delivery time as main constraint and different processing time to process a job. The result of this proposed model shows that the utilization of production machines can be increase with minimal tardiness using fixed delivery time as constraint.

  14. New numerical approach for the modelling of machining applied to aeronautical structural parts

    NASA Astrophysics Data System (ADS)

    Rambaud, Pierrick; Mocellin, Katia

    2018-05-01

    The manufacturing of aluminium alloy structural aerospace parts involves several steps: forming (rolling, forging …etc), heat treatments and machining. Before machining, the manufacturing processes have embedded residual stresses into the workpiece. The final geometry is obtained during this last step, when up to 90% of the raw material volume is removed by machining. During this operation, the mechanical equilibrium of the part is in constant evolution due to the redistribution of the initial stresses. This redistribution is the main cause for workpiece deflections during machining and for distortions - after unclamping. Both may lead to non-conformity of the part regarding the geometrical and dimensional specifications and therefore to rejection of the part or additional conforming steps. In order to improve the machining accuracy and the robustness of the process, the effect of the residual stresses has to be considered for the definition of the machining process plan and even in the geometrical definition of the part. In this paper, the authors present two new numerical approaches concerning the modelling of machining of aeronautical structural parts. The first deals with the use of an immersed volume framework to model the cutting step, improving the robustness and the quality of the resulting mesh compared to the previous version. The second is about the mechanical modelling of the machining problem. The authors thus show that in the framework of rolled aluminium parts the use of a linear elasticity model is functional in the finite element formulation and promising regarding the reduction of computation times.

  15. A Sustainable Model for Integrating Current Topics in Machine Learning Research into the Undergraduate Curriculum

    ERIC Educational Resources Information Center

    Georgiopoulos, M.; DeMara, R. F.; Gonzalez, A. J.; Wu, A. S.; Mollaghasemi, M.; Gelenbe, E.; Kysilka, M.; Secretan, J.; Sharma, C. A.; Alnsour, A. J.

    2009-01-01

    This paper presents an integrated research and teaching model that has resulted from an NSF-funded effort to introduce results of current Machine Learning research into the engineering and computer science curriculum at the University of Central Florida (UCF). While in-depth exposure to current topics in Machine Learning has traditionally occurred…

  16. Abstractive dissociation of oxygen over Al(111): a nonadiabatic quantum model.

    PubMed

    Katz, Gil; Kosloff, Ronnie; Zeiri, Yehuda

    2004-02-22

    The dissociation of oxygen on a clean aluminum surface is studied theoretically. A nonadiabatic quantum dynamical model is used, based on four electronically distinct potential energy surfaces characterized by the extent of charge transfer from the metal to the adsorbate. A flat surface approximation is used to reduce the computation complexity. The conservation of the helicopter angular momentum allows Boltzmann averaging of the outcome of the propagation of a three degrees of freedom wave function. The dissociation event is simulated by solving the time-dependent Schrödinger equation for a period of 30 femtoseconds. As a function of incident kinetic energy, the dissociation yield follows the experimental trend. An attempt at simulation employing only the lowest adiabatic surface failed, qualitatively disagreeing with both experiment and nonadiabatic calculations. The final products, adsorptive dissociation and abstractive dissociation, are obtained by carrying out a semiclassical molecular dynamics simulation with surface hopping which describes the back charge transfer from an oxygen atom negative ion to the surface. The final adsorbed oxygen pair distribution compares well with experiment. By running the dynamical events backward in time, a correlation is established between the products and the initial conditions which lead to their production. Qualitative agreement is thus obtained with recent experiments that show suppression of abstraction by rotational excitation. (c) 2004 American Institute of Physics.

  17. Model studies of hydrogen atom addition and abstraction processes involving ortho-, meta-, and para-benzynes.

    PubMed

    Clark, A E; Davidson, E R

    2001-10-31

    H-atom addition and abstraction processes involving ortho-, meta-, and para-benzyne have been investigated by multiconfigurational self-consistent field methods. The H(A) + H(B)...H(C) reaction (where r(BC) is adjusted to mimic the appropriate singlet-triplet energy gap) is shown to effectively model H-atom addition to benzyne. The doublet multiconfiguration wave functions are shown to mix the "singlet" and "triplet" valence bond structures of H(B)...H(C) along the reaction coordinate; however, the extent of mixing is dependent on the singlet-triplet energy gap (DeltaE(ST)) of the H(B)...H(C) diradical. Early in the reaction, the ground-state wave function is essentially the "singlet" VB function, yet it gains significant "triplet" VB character along the reaction coordinate that allows H(A)-H(B) bond formation. Conversely, the wave function of the first excited state is predominantly the "triplet" VB configuration early in the reaction coordinate, but gains "singlet" VB character when the H-atom is close to a radical center. As a result, the potential energy surface (PES) for H-atom addition to triplet H(B)...H(C) diradical is repulsive! The H3 model predicts, in agreement with the actual calculations on benzyne, that the singlet diradical electrons are not coupled strongly enough to give rise to an activation barrier associated with C-H bond formation. Moreover, this model predicts that the PES for H-atom addition to triplet benzyne will be characterized by a repulsive curve early in the reaction coordinate, followed by a potential avoided crossing with the (pi)1(sigma*)1 state of the phenyl radical. In contrast to H-atom addition, large activation barriers characterize the abstraction process in both the singlet ground state and first triplet state. In the ground state, this barrier results from the weakly avoided crossing of the dominant VB configurations in the ground-state singlet (S0) and first excited singlet (S1) because of the large energy gap between S0

  18. (abstract) Modeling Protein Families and Human Genes: Hidden Markov Models and a Little Beyond

    NASA Technical Reports Server (NTRS)

    Baldi, Pierre

    1994-01-01

    We will first give a brief overview of Hidden Markov Models (HMMs) and their use in Computational Molecular Biology. In particular, we will describe a detailed application of HMMs to the G-Protein-Coupled-Receptor Superfamily. We will also describe a number of analytical results on HMMs that can be used in discrimination tests and database mining. We will then discuss the limitations of HMMs and some new directions of research. We will conclude with some recent results on the application of HMMs to human gene modeling and parsing.

  19. On problems in defining abstract and metaphysical concepts--emergence of a new model.

    PubMed

    Nahod, Bruno; Nahod, Perina Vukša

    2014-12-01

    Basic anthropological terminology is the first project covering terms from the domain of the social sciences under the Croatian Special Field Terminology program (Struna). Problems that have been sporadically noticed or whose existence could have been presumed during the processing of terms mainly from technical fields and sciences have finally emerged in "anthropology". The principles of the General Theory of Terminology (GTT), which are followed in Struna, were put to a truly exacting test, and sometimes stretched beyond their limits when applied to concepts that do not necessarily have references in the physical world; namely, abstract and metaphysical concepts. We are currently developing a new terminographical model based on Idealized Cognitive Models (ICM), which will hopefully ensure a better cross-filed implementation of various types of concepts and their relations. The goal of this paper is to introduce the theoretical bases of our model. Additionally, we will present a pilot study of the series of experiments in which we are trying to investigate the nature of conceptual categorization in special languages and its proposed difference form categorization in general language.

  20. Modelling of human-machine interaction in equipment design of manufacturing cells

    NASA Astrophysics Data System (ADS)

    Cochran, David S.; Arinez, Jorge F.; Collins, Micah T.; Bi, Zhuming

    2017-08-01

    This paper proposes a systematic approach to model human-machine interactions (HMIs) in supervisory control of machining operations; it characterises the coexistence of machines and humans for an enterprise to balance the goals of automation/productivity and flexibility/agility. In the proposed HMI model, an operator is associated with a set of behavioural roles as a supervisor for multiple, semi-automated manufacturing processes. The model is innovative in the sense that (1) it represents an HMI based on its functions for process control but provides the flexibility for ongoing improvements in the execution of manufacturing processes; (2) it provides a computational tool to define functional requirements for an operator in HMIs. The proposed model can be used to design production systems at different levels of an enterprise architecture, particularly at the machine level in a production system where operators interact with semi-automation to accomplish the goal of 'autonomation' - automation that augments the capabilities of human beings.

  1. Association Rule-based Predictive Model for Machine Failure in Industrial Internet of Things

    NASA Astrophysics Data System (ADS)

    Kwon, Jung-Hyok; Lee, Sol-Bee; Park, Jaehoon; Kim, Eui-Jik

    2017-09-01

    This paper proposes an association rule-based predictive model for machine failure in industrial Internet of things (IIoT), which can accurately predict the machine failure in real manufacturing environment by investigating the relationship between the cause and type of machine failure. To develop the predictive model, we consider three major steps: 1) binarization, 2) rule creation, 3) visualization. The binarization step translates item values in a dataset into one or zero, then the rule creation step creates association rules as IF-THEN structures using the Lattice model and Apriori algorithm. Finally, the created rules are visualized in various ways for users’ understanding. An experimental implementation was conducted using R Studio version 3.3.2. The results show that the proposed predictive model realistically predicts machine failure based on association rules.

  2. Behavioral Modeling for Mental Health using Machine Learning Algorithms.

    PubMed

    Srividya, M; Mohanavalli, S; Bhalaji, N

    2018-04-03

    Mental health is an indicator of emotional, psychological and social well-being of an individual. It determines how an individual thinks, feels and handle situations. Positive mental health helps one to work productively and realize their full potential. Mental health is important at every stage of life, from childhood and adolescence through adulthood. Many factors contribute to mental health problems which lead to mental illness like stress, social anxiety, depression, obsessive compulsive disorder, drug addiction, and personality disorders. It is becoming increasingly important to determine the onset of the mental illness to maintain proper life balance. The nature of machine learning algorithms and Artificial Intelligence (AI) can be fully harnessed for predicting the onset of mental illness. Such applications when implemented in real time will benefit the society by serving as a monitoring tool for individuals with deviant behavior. This research work proposes to apply various machine learning algorithms such as support vector machines, decision trees, naïve bayes classifier, K-nearest neighbor classifier and logistic regression to identify state of mental health in a target group. The responses obtained from the target group for the designed questionnaire were first subject to unsupervised learning techniques. The labels obtained as a result of clustering were validated by computing the Mean Opinion Score. These cluster labels were then used to build classifiers to predict the mental health of an individual. Population from various groups like high school students, college students and working professionals were considered as target groups. The research presents an analysis of applying the aforementioned machine learning algorithms on the target groups and also suggests directions for future work.

  3. Rotating magnetizations in electrical machines: Measurements and modeling

    NASA Astrophysics Data System (ADS)

    Thul, Andreas; Steentjes, Simon; Schauerte, Benedikt; Klimczyk, Piotr; Denke, Patrick; Hameyer, Kay

    2018-05-01

    This paper studies the magnetization process in electrical steel sheets for rotational magnetizations as they occur in the magnetic circuit of electrical machines. A four-pole rotational single sheet tester is used to generate the rotating magnetic flux inside the sample. A field-oriented control scheme is implemented to improve the control performance. The magnetization process of different non-oriented materials is analyzed and compared.

  4. Equivalent model of a dually-fed machine for electric drive control systems

    NASA Astrophysics Data System (ADS)

    Ostrovlyanchik, I. Yu; Popolzin, I. Yu

    2018-05-01

    The article shows that the mathematical model of a dually-fed machine is complicated because of the presence of a controlled voltage source in the rotor circuit. As a method of obtaining a mathematical model, the method of a generalized two-phase electric machine is applied and a rotating orthogonal coordinate system is chosen that is associated with the representing vector of a stator current. In the chosen coordinate system in the operator form the differential equations of electric equilibrium for the windings of the generalized machine (the Kirchhoff equation) are written together with the expression for the moment, which determines the electromechanical energy transformation in the machine. Equations are transformed so that they connect the currents of the windings, that determine the moment of the machine, and the voltages on these windings. The structural diagram of the machine is assigned to the written equations. Based on the written equations and accepted assumptions, expressions were obtained for the balancing the EMF of windings, and on the basis of these expressions an equivalent mathematical model of a dually-fed machine is proposed, convenient for use in electric drive control systems.

  5. A comparison of machine learning and Bayesian modelling for molecular serotyping.

    PubMed

    Newton, Richard; Wernisch, Lorenz

    2017-08-11

    Streptococcus pneumoniae is a human pathogen that is a major cause of infant mortality. Identifying the pneumococcal serotype is an important step in monitoring the impact of vaccines used to protect against disease. Genomic microarrays provide an effective method for molecular serotyping. Previously we developed an empirical Bayesian model for the classification of serotypes from a molecular serotyping array. With only few samples available, a model driven approach was the only option. In the meanwhile, several thousand samples have been made available to us, providing an opportunity to investigate serotype classification by machine learning methods, which could complement the Bayesian model. We compare the performance of the original Bayesian model with two machine learning algorithms: Gradient Boosting Machines and Random Forests. We present our results as an example of a generic strategy whereby a preliminary probabilistic model is complemented or replaced by a machine learning classifier once enough data are available. Despite the availability of thousands of serotyping arrays, a problem encountered when applying machine learning methods is the lack of training data containing mixtures of serotypes; due to the large number of possible combinations. Most of the available training data comprises samples with only a single serotype. To overcome the lack of training data we implemented an iterative analysis, creating artificial training data of serotype mixtures by combining raw data from single serotype arrays. With the enhanced training set the machine learning algorithms out perform the original Bayesian model. However, for serotypes currently lacking sufficient training data the best performing implementation was a combination of the results of the Bayesian Model and the Gradient Boosting Machine. As well as being an effective method for classifying biological data, machine learning can also be used as an efficient method for revealing subtle biological

  6. Improving protein–protein interactions prediction accuracy using protein evolutionary information and relevance vector machine model

    PubMed Central

    An, Ji‐Yong; Meng, Fan‐Rong; Chen, Xing; Yan, Gui‐Ying; Hu, Ji‐Pu

    2016-01-01

    Abstract Predicting protein–protein interactions (PPIs) is a challenging task and essential to construct the protein interaction networks, which is important for facilitating our understanding of the mechanisms of biological systems. Although a number of high‐throughput technologies have been proposed to predict PPIs, there are unavoidable shortcomings, including high cost, time intensity, and inherently high false positive rates. For these reasons, many computational methods have been proposed for predicting PPIs. However, the problem is still far from being solved. In this article, we propose a novel computational method called RVM‐BiGP that combines the relevance vector machine (RVM) model and Bi‐gram Probabilities (BiGP) for PPIs detection from protein sequences. The major improvement includes (1) Protein sequences are represented using the Bi‐gram probabilities (BiGP) feature representation on a Position Specific Scoring Matrix (PSSM), in which the protein evolutionary information is contained; (2) For reducing the influence of noise, the Principal Component Analysis (PCA) method is used to reduce the dimension of BiGP vector; (3) The powerful and robust Relevance Vector Machine (RVM) algorithm is used for classification. Five‐fold cross‐validation experiments executed on yeast and Helicobacter pylori datasets, which achieved very high accuracies of 94.57 and 90.57%, respectively. Experimental results are significantly better than previous methods. To further evaluate the proposed method, we compare it with the state‐of‐the‐art support vector machine (SVM) classifier on the yeast dataset. The experimental results demonstrate that our RVM‐BiGP method is significantly better than the SVM‐based method. In addition, we achieved 97.15% accuracy on imbalance yeast dataset, which is higher than that of balance yeast dataset. The promising experimental results show the efficiency and robust of the proposed method, which can be an automatic

  7. Neural networks with fuzzy Petri nets for modeling a machining process

    NASA Astrophysics Data System (ADS)

    Hanna, Moheb M.

    1998-03-01

    The paper presents an intelligent architecture based a feedforward neural network with fuzzy Petri nets for modeling product quality in a CNC machining center. It discusses how the proposed architecture can be used for modeling, monitoring and control a product quality specification such as surface roughness. The surface roughness represents the output quality specification manufactured by a CNC machining center as a result of a milling process. The neural network approach employed the selected input parameters which defined by the machine operator via the CNC code. The fuzzy Petri nets approach utilized the exact input milling parameters, such as spindle speed, feed rate, tool diameter and coolant (off/on), which can be obtained via the machine or sensors system. An aim of the proposed architecture is to model the demanded quality of surface roughness as high, medium or low.

  8. Human factors model concerning the man-machine interface of mining crewstations

    NASA Technical Reports Server (NTRS)

    Rider, James P.; Unger, Richard L.

    1989-01-01

    The U.S. Bureau of Mines is developing a computer model to analyze the human factors aspect of mining machine operator compartments. The model will be used as a research tool and as a design aid. It will have the capability to perform the following: simulated anthropometric or reach assessment, visibility analysis, illumination analysis, structural analysis of the protective canopy, operator fatigue analysis, and computation of an ingress-egress rating. The model will make extensive use of graphics to simplify data input and output. Two dimensional orthographic projections of the machine and its operator compartment are digitized and the data rebuilt into a three dimensional representation of the mining machine. Anthropometric data from either an individual or any size population may be used. The model is intended for use by equipment manufacturers and mining companies during initial design work on new machines. In addition to its use in machine design, the model should prove helpful as an accident investigation tool and for determining the effects of machine modifications made in the field on the critical areas of visibility and control reach ability.

  9. 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.

  10. Vacation model for Markov machine repair problem with two heterogeneous unreliable servers and threshold recovery

    NASA Astrophysics Data System (ADS)

    Jain, Madhu; Meena, Rakesh Kumar

    2018-03-01

    Markov model of multi-component machining system comprising two unreliable heterogeneous servers and mixed type of standby support has been studied. The repair job of broken down machines is done on the basis of bi-level threshold policy for the activation of the servers. The server returns back to render repair job when the pre-specified workload of failed machines is build up. The first (second) repairman turns on only when the work load of N1 (N2) failed machines is accumulated in the system. The both servers may go for vacation in case when all the machines are in good condition and there are no pending repair jobs for the repairmen. Runge-Kutta method is implemented to solve the set of governing equations used to formulate the Markov model. Various system metrics including the mean queue length, machine availability, throughput, etc., are derived to determine the performance of the machining system. To provide the computational tractability of the present investigation, a numerical illustration is provided. A cost function is also constructed to determine the optimal repair rate of the server by minimizing the expected cost incurred on the system. The hybrid soft computing method is considered to develop the adaptive neuro-fuzzy inference system (ANFIS). The validation of the numerical results obtained by Runge-Kutta approach is also facilitated by computational results generated by ANFIS.

  11. Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary View.

    PubMed

    Luo, Wei; Phung, Dinh; Tran, Truyen; Gupta, Sunil; Rana, Santu; Karmakar, Chandan; Shilton, Alistair; Yearwood, John; Dimitrova, Nevenka; Ho, Tu Bao; Venkatesh, Svetha; Berk, Michael

    2016-12-16

    As more and more researchers are turning to big data for new opportunities of biomedical discoveries, machine learning models, as the backbone of big data analysis, are mentioned more often in biomedical journals. However, owing to the inherent complexity of machine learning methods, they are prone to misuse. Because of the flexibility in specifying machine learning models, the results are often insufficiently reported in research articles, hindering reliable assessment of model validity and consistent interpretation of model outputs. To attain a set of guidelines on the use of machine learning predictive models within clinical settings to make sure the models are correctly applied and sufficiently reported so that true discoveries can be distinguished from random coincidence. A multidisciplinary panel of machine learning experts, clinicians, and traditional statisticians were interviewed, using an iterative process in accordance with the Delphi method. The process produced a set of guidelines that consists of (1) a list of reporting items to be included in a research article and (2) a set of practical sequential steps for developing predictive models. A set of guidelines was generated to enable correct application of machine learning models and consistent reporting of model specifications and results in biomedical research. We believe that such guidelines will accelerate the adoption of big data analysis, particularly with machine learning methods, in the biomedical research community. ©Wei Luo, Dinh Phung, Truyen Tran, Sunil Gupta, Santu Rana, Chandan Karmakar, Alistair Shilton, John Yearwood, Nevenka Dimitrova, Tu Bao Ho, Svetha Venkatesh, Michael Berk. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 16.12.2016.

  12. A Wavelet Support Vector Machine Combination Model for Singapore Tourist Arrival to Malaysia

    NASA Astrophysics Data System (ADS)

    Rafidah, A.; Shabri, Ani; Nurulhuda, A.; Suhaila, Y.

    2017-08-01

    In this study, wavelet support vector machine model (WSVM) is proposed and applied for monthly data Singapore tourist time series prediction. The WSVM model is combination between wavelet analysis and support vector machine (SVM). In this study, we have two parts, first part we compare between the kernel function and second part we compare between the developed models with single model, SVM. The result showed that kernel function linear better than RBF while WSVM outperform with single model SVM to forecast monthly Singapore tourist arrival to Malaysia.

  13. Quantum turing machine and brain model represented by Fock space

    NASA Astrophysics Data System (ADS)

    Iriyama, Satoshi; Ohya, Masanori

    2016-05-01

    The adaptive dynamics is known as a new mathematics to treat with a complex phenomena, for example, chaos, quantum algorithm and psychological phenomena. In this paper, we briefly review the notion of the adaptive dynamics, and explain the definition of the generalized Turing machine (GTM) and recognition process represented by the Fock space. Moreover, we show that there exists the quantum channel which is described by the GKSL master equation to achieve the Chaos Amplifier used in [M. Ohya and I. V. Volovich, J. Opt. B 5(6) (2003) 639., M. Ohya and I. V. Volovich, Rep. Math. Phys. 52(1) (2003) 25.

  14. Computational Model for Impact-Resisting Critical Thickness of High-Speed Machine Outer Protective Plate

    NASA Astrophysics Data System (ADS)

    Wu, Huaying; Wang, Li Zhong; Wang, Yantao; Yuan, Xiaolei

    2018-05-01

    The blade or surface grinding blade of the hypervelocity grinding wheel may be damaged due to too high rotation rate of the spindle of the machine and then fly out. Its speed as a projectile may severely endanger the field persons. Critical thickness model of the protective plate of the high-speed machine is studied in this paper. For easy analysis, the shapes of the possible impact objects flying from the high-speed machine are simplified as sharp-nose model, ball-nose model and flat-nose model. Whose front ending shape to represent point, line and surface contacting. Impact analysis based on J-C model is performed for the low-carbon steel plate with different thicknesses in this paper. One critical thickness computational model for the protective plate of high-speed machine is established according to the damage characteristics of the thin plate to get relation among plate thickness and mass, shape and size and impact speed of impact object. The air cannon is used for impact test. The model accuracy is validated. This model can guide identification of the thickness of single-layer outer protective plate of a high-speed machine.

  15. Modeling of heat transfer in compacted machining chips during friction consolidation process

    NASA Astrophysics Data System (ADS)

    Abbas, Naseer; Deng, Xiaomin; Li, Xiao; Reynolds, Anthony

    2018-04-01

    The current study aims to provide an understanding of the heat transfer process in compacted aluminum alloy AA6061 machining chips during the friction consolidation process (FCP) through experimental investigations and mathematical modelling and numerical simulation. Compaction and friction consolidation of machining chips is the first stage of the Friction Extrusion Process (FEP), which is a novel method for recycling machining chips to produce useful products such as wires. In this study, compacted machining chips are modelled as a continuum whose material properties vary with density during friction consolidation. Based on density and temperature dependent thermal properties, the temperature field in the chip material and process chamber caused by frictional heating during the friction consolidation process is predicted. The predicted temperature field is found to compare well with temperature measurements at select points where such measurements can be made using thermocouples.

  16. Thermal Error Test and Intelligent Modeling Research on the Spindle of High Speed CNC Machine Tools

    NASA Astrophysics Data System (ADS)

    Luo, Zhonghui; Peng, Bin; Xiao, Qijun; Bai, Lu

    2018-03-01

    Thermal error is the main factor affecting the accuracy of precision machining. Through experiments, this paper studies the thermal error test and intelligent modeling for the spindle of vertical high speed CNC machine tools in respect of current research focuses on thermal error of machine tool. Several testing devices for thermal error are designed, of which 7 temperature sensors are used to measure the temperature of machine tool spindle system and 2 displacement sensors are used to detect the thermal error displacement. A thermal error compensation model, which has a good ability in inversion prediction, is established by applying the principal component analysis technology, optimizing the temperature measuring points, extracting the characteristic values closely associated with the thermal error displacement, and using the artificial neural network technology.

  17. A Framework for Modeling Human-Machine Interactions

    NASA Technical Reports Server (NTRS)

    Shafto, Michael G.; Rosekind, Mark R. (Technical Monitor)

    1996-01-01

    Modern automated flight-control systems employ a variety of different behaviors, or modes, for managing the flight. While developments in cockpit automation have resulted in workload reduction and economical advantages, they have also given rise to an ill-defined class of human-machine problems, sometimes referred to as 'automation surprises'. Our interest in applying formal methods for describing human-computer interaction stems from our ongoing research on cockpit automation. In this area of aeronautical human factors, there is much concern about how flight crews interact with automated flight-control systems, so that the likelihood of making errors, in particular mode-errors, is minimized and the consequences of such errors are contained. The goal of the ongoing research on formal methods in this context is: (1) to develop a framework for describing human interaction with control systems; (2) to formally categorize such automation surprises; and (3) to develop tests for identification of these categories early in the specification phase of a new human-machine system.

  18. Spindle Thermal Error Optimization Modeling of a Five-axis Machine Tool

    NASA Astrophysics Data System (ADS)

    Guo, Qianjian; Fan, Shuo; Xu, Rufeng; Cheng, Xiang; Zhao, Guoyong; Yang, Jianguo

    2017-05-01

    Aiming at the problem of low machining accuracy and uncontrollable thermal errors of NC machine tools, spindle thermal error measurement, modeling and compensation of a two turntable five-axis machine tool are researched. Measurement experiment of heat sources and thermal errors are carried out, and GRA(grey relational analysis) method is introduced into the selection of temperature variables used for thermal error modeling. In order to analyze the influence of different heat sources on spindle thermal errors, an ANN (artificial neural network) model is presented, and ABC(artificial bee colony) algorithm is introduced to train the link weights of ANN, a new ABC-NN(Artificial bee colony-based neural network) modeling method is proposed and used in the prediction of spindle thermal errors. In order to test the prediction performance of ABC-NN model, an experiment system is developed, the prediction results of LSR (least squares regression), ANN and ABC-NN are compared with the measurement results of spindle thermal errors. Experiment results show that the prediction accuracy of ABC-NN model is higher than LSR and ANN, and the residual error is smaller than 3 μm, the new modeling method is feasible. The proposed research provides instruction to compensate thermal errors and improve machining accuracy of NC machine tools.

  19. Modelling of internal architecture of kinesin nanomotor as a machine language.

    PubMed

    Khataee, H R; Ibrahim, M Y

    2012-09-01

    Kinesin is a protein-based natural nanomotor that transports molecular cargoes within cells by walking along microtubules. Kinesin nanomotor is considered as a bio-nanoagent which is able to sense the cell through its sensors (i.e. its heads and tail), make the decision internally and perform actions on the cell through its actuator (i.e. its motor domain). The study maps the agent-based architectural model of internal decision-making process of kinesin nanomotor to a machine language using an automata algorithm. The applied automata algorithm receives the internal agent-based architectural model of kinesin nanomotor as a deterministic finite automaton (DFA) model and generates a regular machine language. The generated regular machine language was acceptable by the architectural DFA model of the nanomotor and also in good agreement with its natural behaviour. The internal agent-based architectural model of kinesin nanomotor indicates the degree of autonomy and intelligence of the nanomotor interactions with its cell. Thus, our developed regular machine language can model the degree of autonomy and intelligence of kinesin nanomotor interactions with its cell as a language. Modelling of internal architectures of autonomous and intelligent bio-nanosystems as machine languages can lay the foundation towards the concept of bio-nanoswarms and next phases of the bio-nanorobotic systems development.

  20. Machine Learning Algorithms Outperform Conventional Regression Models in Predicting Development of Hepatocellular Carcinoma

    PubMed Central

    Singal, Amit G.; Mukherjee, Ashin; Elmunzer, B. Joseph; Higgins, Peter DR; Lok, Anna S.; Zhu, Ji; Marrero, Jorge A; Waljee, Akbar K

    2015-01-01

    Background Predictive models for hepatocellular carcinoma (HCC) have been limited by modest accuracy and lack of validation. Machine learning algorithms offer a novel methodology, which may improve HCC risk prognostication among patients with cirrhosis. Our study's aim was to develop and compare predictive models for HCC development among cirrhotic patients, using conventional regression analysis and machine learning algorithms. Methods We enrolled 442 patients with Child A or B cirrhosis at the University of Michigan between January 2004 and September 2006 (UM cohort) and prospectively followed them until HCC development, liver transplantation, death, or study termination. Regression analysis and machine learning algorithms were used to construct predictive models for HCC development, which were tested on an independent validation cohort from the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial. Both models were also compared to the previously published HALT-C model. Discrimination was assessed using receiver operating characteristic curve analysis and diagnostic accuracy was assessed with net reclassification improvement and integrated discrimination improvement statistics. Results After a median follow-up of 3.5 years, 41 patients developed HCC. The UM regression model had a c-statistic of 0.61 (95%CI 0.56-0.67), whereas the machine learning algorithm had a c-statistic of 0.64 (95%CI 0.60–0.69) in the validation cohort. The machine learning algorithm had significantly better diagnostic accuracy as assessed by net reclassification improvement (p<0.001) and integrated discrimination improvement (p=0.04). The HALT-C model had a c-statistic of 0.60 (95%CI 0.50-0.70) in the validation cohort and was outperformed by the machine learning algorithm (p=0.047). Conclusion Machine learning algorithms improve the accuracy of risk stratifying patients with cirrhosis and can be used to accurately identify patients at high-risk for developing HCC

  1. An Introduction to Topic Modeling as an Unsupervised Machine Learning Way to Organize Text Information

    ERIC Educational Resources Information Center

    Snyder, Robin M.

    2015-01-01

    The field of topic modeling has become increasingly important over the past few years. Topic modeling is an unsupervised machine learning way to organize text (or image or DNA, etc.) information such that related pieces of text can be identified. This paper/session will present/discuss the current state of topic modeling, why it is important, and…

  2. Temperature control of fimbriation circuit switch in uropathogenic Escherichia coli: quantitative analysis via automated model abstraction.

    PubMed

    Kuwahara, Hiroyuki; Myers, Chris J; Samoilov, Michael S

    2010-03-26

    Uropathogenic Escherichia coli (UPEC) represent the predominant cause of urinary tract infections (UTIs). A key UPEC molecular virulence mechanism is type 1 fimbriae, whose expression is controlled by the orientation of an invertible chromosomal DNA element-the fim switch. Temperature has been shown to act as a major regulator of fim switching behavior and is overall an important indicator as well as functional feature of many urologic diseases, including UPEC host-pathogen interaction dynamics. Given this panoptic physiological role of temperature during UTI progression and notable empirical challenges to its direct in vivo studies, in silico modeling of corresponding biochemical and biophysical mechanisms essential to UPEC pathogenicity may significantly aid our understanding of the underlying disease processes. However, rigorous computational analysis of biological systems, such as fim switch temperature control circuit, has hereto presented a notoriously demanding problem due to both the substantial complexity of the gene regulatory networks involved as well as their often characteristically discrete and stochastic dynamics. To address these issues, we have developed an approach that enables automated multiscale abstraction of biological system descriptions based on reaction kinetics. Implemented as a computational tool, this method has allowed us to efficiently analyze the modular organization and behavior of the E. coli fimbriation switch circuit at different temperature settings, thus facilitating new insights into this mode of UPEC molecular virulence regulation. In particular, our results suggest that, with respect to its role in shutting down fimbriae expression, the primary function of FimB recombinase may be to effect a controlled down-regulation (rather than increase) of the ON-to-OFF fim switching rate via temperature-dependent suppression of competing dynamics mediated by recombinase FimE. Our computational analysis further implies that this down

  3. Machine Visual Motion Detection Modeled on Vertebrate Retina

    DTIC Science & Technology

    1988-01-01

    18. NUMBER OF PAGES 9 19a. NAME OF RESPONSIBLE PERSON a. REPORT unclassified b . ABSTRACT unclassified c. THIS PAGE unclassified Standard Form...mechanism of direction selectivity. (a) shows the use of persistent lateral inhibition to block conduction in the null direction. ( b ) shows the use of...Bipolar elements Bipolar ( B ) elements compare the inputs from local receptor and horizontal elements, passing on the positive value of the difference

  4. A hybrid analytical model for open-circuit field calculation of multilayer interior permanent magnet machines

    NASA Astrophysics Data System (ADS)

    Zhang, Zhen; Xia, Changliang; Yan, Yan; Geng, Qiang; Shi, Tingna

    2017-08-01

    Due to the complicated rotor structure and nonlinear saturation of rotor bridges, it is difficult to build a fast and accurate analytical field calculation model for multilayer interior permanent magnet (IPM) machines. In this paper, a hybrid analytical model suitable for the open-circuit field calculation of multilayer IPM machines is proposed by coupling the magnetic equivalent circuit (MEC) method and the subdomain technique. In the proposed analytical model, the rotor magnetic field is calculated by the MEC method based on the Kirchhoff's law, while the field in the stator slot, slot opening and air-gap is calculated by subdomain technique based on the Maxwell's equation. To solve the whole field distribution of the multilayer IPM machines, the coupled boundary conditions on the rotor surface are deduced for the coupling of the rotor MEC and the analytical field distribution of the stator slot, slot opening and air-gap. The hybrid analytical model can be used to calculate the open-circuit air-gap field distribution, back electromotive force (EMF) and cogging torque of multilayer IPM machines. Compared with finite element analysis (FEA), it has the advantages of faster modeling, less computation source occupying and shorter time consuming, and meanwhile achieves the approximate accuracy. The analytical model is helpful and applicable for the open-circuit field calculation of multilayer IPM machines with any size and pole/slot number combination.

  5. Product Quality Modelling Based on Incremental Support Vector Machine

    NASA Astrophysics Data System (ADS)

    Wang, J.; Zhang, W.; Qin, B.; Shi, W.

    2012-05-01

    Incremental Support vector machine (ISVM) is a new learning method developed in recent years based on the foundations of statistical learning theory. It is suitable for the problem of sequentially arriving field data and has been widely used for product quality prediction and production process optimization. However, the traditional ISVM learning does not consider the quality of the incremental data which may contain noise and redundant data; it will affect the learning speed and accuracy to a great extent. In order to improve SVM training speed and accuracy, a modified incremental support vector machine (MISVM) is proposed in this paper. Firstly, the margin vectors are extracted according to the Karush-Kuhn-Tucker (KKT) condition; then the distance from the margin vectors to the final decision hyperplane is calculated to evaluate the importance of margin vectors, where the margin vectors are removed while their distance exceed the specified value; finally, the original SVs and remaining margin vectors are used to update the SVM. The proposed MISVM can not only eliminate the unimportant samples such as noise samples, but also can preserve the important samples. The MISVM has been experimented on two public data and one field data of zinc coating weight in strip hot-dip galvanizing, and the results shows that the proposed method can improve the prediction accuracy and the training speed effectively. Furthermore, it can provide the necessary decision supports and analysis tools for auto control of product quality, and also can extend to other process industries, such as chemical process and manufacturing process.

  6. Design and finite element analysis of micro punch CNC machine modeling for medical devices

    NASA Astrophysics Data System (ADS)

    Pranoto, Sigiet Haryo; Mahardika, Muslim

    2018-03-01

    Research on micromanufacturing has been conducted. Miniaturization and weight reduction of various industrial products continue to be developed, machines with high accuracy and good quality of machining results are needed recently. This research includes design and simulation of Micro Punch CNC Machine using Abaqus with pneumatic system. This article concern of modeling simulation of punching miniplate titanium with 0.6 MPa of pressure and 500 µm of thickness. This study explaining von misses stress, safety factor and displacement analysis while the machine had the load of punching. The result gives the reaction forced of punching is 0.5 MPa on punch tip and maximum displacement is 3.237 × 10-1 mm. The safety factor is over than 12, and considered it safe for manufacturing process.

  7. Progressive sampling-based Bayesian optimization for efficient and automatic machine learning model selection.

    PubMed

    Zeng, Xueqiang; Luo, Gang

    2017-12-01

    Machine learning is broadly used for clinical data analysis. Before training a model, a machine learning algorithm must be selected. Also, the values of one or more model parameters termed hyper-parameters must be set. Selecting algorithms and hyper-parameter values requires advanced machine learning knowledge and many labor-intensive manual iterations. To lower the bar to machine learning, miscellaneous automatic selection methods for algorithms and/or hyper-parameter values have been proposed. Existing automatic selection methods are inefficient on large data sets. This poses a challenge for using machine learning in the clinical big data era. To address the challenge, this paper presents progressive sampling-based Bayesian optimization, an efficient and automatic selection method for both algorithms and hyper-parameter values. We report an implementation of the method. We show that compared to a state of the art automatic selection method, our method can significantly reduce search time, classification error rate, and standard deviation of error rate due to randomization. This is major progress towards enabling fast turnaround in identifying high-quality solutions required by many machine learning-based clinical data analysis tasks.

  8. Hybrid forecasting of chaotic processes: Using machine learning in conjunction with a knowledge-based model

    NASA Astrophysics Data System (ADS)

    Pathak, Jaideep; Wikner, Alexander; Fussell, Rebeckah; Chandra, Sarthak; Hunt, Brian R.; Girvan, Michelle; Ott, Edward

    2018-04-01

    A model-based approach to forecasting chaotic dynamical systems utilizes knowledge of the mechanistic processes governing the dynamics to build an approximate mathematical model of the system. In contrast, machine learning techniques have demonstrated promising results for forecasting chaotic systems purely from past time series measurements of system state variables (training data), without prior knowledge of the system dynamics. The motivation for this paper is the potential of machine learning for filling in the gaps in our underlying mechanistic knowledge that cause widely-used knowledge-based models to be inaccurate. Thus, we here propose a general method that leverages the advantages of these two approaches by combining a knowledge-based model and a machine learning technique to build a hybrid forecasting scheme. Potential applications for such an approach are numerous (e.g., improving weather forecasting). We demonstrate and test the utility of this approach using a particular illustrative version of a machine learning known as reservoir computing, and we apply the resulting hybrid forecaster to a low-dimensional chaotic system, as well as to a high-dimensional spatiotemporal chaotic system. These tests yield extremely promising results in that our hybrid technique is able to accurately predict for a much longer period of time than either its machine-learning component or its model-based component alone.

  9. Interpreting linear support vector machine models with heat map molecule coloring

    PubMed Central

    2011-01-01

    Background Model-based virtual screening plays an important role in the early drug discovery stage. The outcomes of high-throughput screenings are a valuable source for machine learning algorithms to infer such models. Besides a strong performance, the interpretability of a machine learning model is a desired property to guide the optimization of a compound in later drug discovery stages. Linear support vector machines showed to have a convincing performance on large-scale data sets. The goal of this study is to present a heat map molecule coloring technique to interpret linear support vector machine models. Based on the weights of a linear model, the visualization approach colors each atom and bond of a compound according to its importance for activity. Results We evaluated our approach on a toxicity data set, a chromosome aberration data set, and the maximum unbiased validation data sets. The experiments show that our method sensibly visualizes structure-property and structure-activity relationships of a linear support vector machine model. The coloring of ligands in the binding pocket of several crystal structures of a maximum unbiased validation data set target indicates that our approach assists to determine the correct ligand orientation in the binding pocket. Additionally, the heat map coloring enables the identification of substructures important for the binding of an inhibitor. Conclusions In combination with heat map coloring, linear support vector machine models can help to guide the modification of a compound in later stages of drug discovery. Particularly substructures identified as important by our method might be a starting point for optimization of a lead compound. The heat map coloring should be considered as complementary to structure based modeling approaches. As such, it helps to get a better understanding of the binding mode of an inhibitor. PMID:21439031

  10. Automating Construction of Machine Learning Models With Clinical Big Data: Proposal Rationale and Methods.

    PubMed

    Luo, Gang; Stone, Bryan L; Johnson, Michael D; Tarczy-Hornoch, Peter; Wilcox, Adam B; Mooney, Sean D; Sheng, Xiaoming; Haug, Peter J; Nkoy, Flory L

    2017-08-29

    To improve health outcomes and cut health care costs, we often need to conduct prediction/classification using large clinical datasets (aka, clinical big data), for example, to identify high-risk patients for preventive interventions. Machine learning has been proposed as a key technology for doing this. Machine learning has won most data science competitions and could support many clinical activities, yet only 15% of hospitals use it for even limited purposes. Despite familiarity with data, health care researchers often lack machine learning expertise to directly use clinical big data, creating a hurdle in realizing value from their data. Health care researchers can work with data scientists with deep machine learning knowledge, but it takes time and effort for both parties to communicate effectively. Facing a shortage in the United States of data scientists and hiring competition from companies with deep pockets, health care systems have difficulty recruiting data scientists. Building and generalizing a machine learning model often requires hundreds to thousands of manual iterations by data scientists to select the following: (1) hyper-parameter values and complex algorithms that greatly affect model accuracy and (2) operators and periods for temporally aggregating clinical attributes (eg, whether a patient's weight kept rising in the past year). This process becomes infeasible with limited budgets. This study's goal is to enable health care researchers to directly use clinical big data, make machine learning feasible with limited budgets and data scientist resources, and realize value from data. This study will allow us to achieve the following: (1) finish developing the new software, Automated Machine Learning (Auto-ML), to automate model selection for machine learning with clinical big data and validate Auto-ML on seven benchmark modeling problems of clinical importance; (2) apply Auto-ML and novel methodology to two new modeling problems crucial for care

  11. Automating Construction of Machine Learning Models With Clinical Big Data: Proposal Rationale and Methods

    PubMed Central

    Stone, Bryan L; Johnson, Michael D; Tarczy-Hornoch, Peter; Wilcox, Adam B; Mooney, Sean D; Sheng, Xiaoming; Haug, Peter J; Nkoy, Flory L

    2017-01-01

    Background To improve health outcomes and cut health care costs, we often need to conduct prediction/classification using large clinical datasets (aka, clinical big data), for example, to identify high-risk patients for preventive interventions. Machine learning has been proposed as a key technology for doing this. Machine learning has won most data science competitions and could support many clinical activities, yet only 15% of hospitals use it for even limited purposes. Despite familiarity with data, health care researchers often lack machine learning expertise to directly use clinical big data, creating a hurdle in realizing value from their data. Health care researchers can work with data scientists with deep machine learning knowledge, but it takes time and effort for both parties to communicate effectively. Facing a shortage in the United States of data scientists and hiring competition from companies with deep pockets, health care systems have difficulty recruiting data scientists. Building and generalizing a machine learning model often requires hundreds to thousands of manual iterations by data scientists to select the following: (1) hyper-parameter values and complex algorithms that greatly affect model accuracy and (2) operators and periods for temporally aggregating clinical attributes (eg, whether a patient’s weight kept rising in the past year). This process becomes infeasible with limited budgets. Objective This study’s goal is to enable health care researchers to directly use clinical big data, make machine learning feasible with limited budgets and data scientist resources, and realize value from data. Methods This study will allow us to achieve the following: (1) finish developing the new software, Automated Machine Learning (Auto-ML), to automate model selection for machine learning with clinical big data and validate Auto-ML on seven benchmark modeling problems of clinical importance; (2) apply Auto-ML and novel methodology to two new

  12. Colour Model for Outdoor Machine Vision for Tropical Regions and its Comparison with the CIE Model

    NASA Astrophysics Data System (ADS)

    Sahragard, Nasrolah; Ramli, Abdul Rahman B.; Hamiruce Marhaban, Mohammad; Mansor, Shattri B.

    2011-02-01

    Accurate modeling of daylight and surface reflectance are very useful for most outdoor machine vision applications specifically those which are based on color recognition. Existing daylight CIE model has drawbacks that limit its ability to predict the color of incident light. These limitations include lack of considering ambient light, effects of light reflected off the ground, and context specific information. Previously developed color model is only tested for a few geographical places in North America and its accountability is under question for other places in the world. Besides, existing surface reflectance models are not easily applied to outdoor images. A reflectance model with combined diffuse and specular reflection in normalized HSV color space could be used to predict color. In this paper, a new daylight color model showing the color of daylight for a broad range of sky conditions is developed which will suit weather conditions of tropical places such as Malaysia. A comparison of this daylight color model and daylight CIE model will be discussed. The colors of matte and specular surfaces have been estimated by use of the developed color model and surface reflection function in this paper. The results are shown to be highly reliable.

  13. Modeling human-machine interactions for operations room layouts

    NASA Astrophysics Data System (ADS)

    Hendy, Keith C.; Edwards, Jack L.; Beevis, David

    2000-11-01

    The LOCATE layout analysis tool was used to analyze three preliminary configurations for the Integrated Command Environment (ICE) of a future USN platform. LOCATE develops a cost function reflecting the quality of all human-human and human-machine communications within a workspace. This proof- of-concept study showed little difference between the efficacy of the preliminary designs selected for comparison. This was thought to be due to the limitations of the study, which included the assumption of similar size for each layout and a lack of accurate measurement data for various objects in the designs, due largely to their notional nature. Based on these results, the USN offered an opportunity to conduct a LOCATE analysis using more appropriate assumptions. A standard crew was assumed, and subject matter experts agreed on the communications patterns for the analysis. Eight layouts were evaluated with the concepts of coordination and command factored into the analysis. Clear differences between the layouts emerged. The most promising design was refined further by the USN, and a working mock-up built for human-in-the-loop evaluation. LOCATE was applied to this configuration for comparison with the earlier analyses.

  14. PredicT-ML: a tool for automating machine learning model building with big clinical data.

    PubMed

    Luo, Gang

    2016-01-01

    Predictive modeling is fundamental to transforming large clinical data sets, or "big clinical data," into actionable knowledge for various healthcare applications. Machine learning is a major predictive modeling approach, but two barriers make its use in healthcare challenging. First, a machine learning tool user must choose an algorithm and assign one or more model parameters called hyper-parameters before model training. The algorithm and hyper-parameter values used typically impact model accuracy by over 40 %, but their selection requires many labor-intensive manual iterations that can be difficult even for computer scientists. Second, many clinical attributes are repeatedly recorded over time, requiring temporal aggregation before predictive modeling can be performed. Many labor-intensive manual iterations are required to identify a good pair of aggregation period and operator for each clinical attribute. Both barriers result in time and human resource bottlenecks, and preclude healthcare administrators and researchers from asking a series of what-if questions when probing opportunities to use predictive models to improve outcomes and reduce costs. This paper describes our design of and vision for PredicT-ML (prediction tool using machine learning), a software system that aims to overcome these barriers and automate machine learning model building with big clinical data. The paper presents the detailed design of PredicT-ML. PredicT-ML will open the use of big clinical data to thousands of healthcare administrators and researchers and increase the ability to advance clinical research and improve healthcare.

  15. Quantitative approaches to energy and glucose homeostasis: machine learning and modelling for precision understanding and prediction

    PubMed Central

    Murphy, Kevin G.; Jones, Nick S.

    2018-01-01

    Obesity is a major global public health problem. Understanding how energy homeostasis is regulated, and can become dysregulated, is crucial for developing new treatments for obesity. Detailed recording of individual behaviour and new imaging modalities offer the prospect of medically relevant models of energy homeostasis that are both understandable and individually predictive. The profusion of data from these sources has led to an interest in applying machine learning techniques to gain insight from these large, relatively unstructured datasets. We review both physiological models and machine learning results across a diverse range of applications in energy homeostasis, and highlight how modelling and machine learning can work together to improve predictive ability. We collect quantitative details in a comprehensive mathematical supplement. We also discuss the prospects of forecasting homeostatic behaviour and stress the importance of characterizing stochasticity within and between individuals in order to provide practical, tailored forecasts and guidance to combat the spread of obesity. PMID:29367240

  16. Modelling of Tool Wear and Residual Stress during Machining of AISI H13 Tool Steel

    NASA Astrophysics Data System (ADS)

    Outeiro, José C.; Umbrello, Domenico; Pina, José C.; Rizzuti, Stefania

    2007-05-01

    Residual stresses can enhance or impair the ability of a component to withstand loading conditions in service (fatigue, creep, stress corrosion cracking, etc.), depending on their nature: compressive or tensile, respectively. This poses enormous problems in structural assembly as this affects the structural integrity of the whole part. In addition, tool wear issues are of critical importance in manufacturing since these affect component quality, tool life and machining cost. Therefore, prediction and control of both tool wear and the residual stresses in machining are absolutely necessary. In this work, a two-dimensional Finite Element model using an implicit Lagrangian formulation with an automatic remeshing was applied to simulate the orthogonal cutting process of AISI H13 tool steel. To validate such model the predicted and experimentally measured chip geometry, cutting forces, temperatures, tool wear and residual stresses on the machined affected layers were compared. The proposed FE model allowed us to investigate the influence of tool geometry, cutting regime parameters and tool wear on residual stress distribution in the machined surface and subsurface of AISI H13 tool steel. The obtained results permit to conclude that in order to reduce the magnitude of surface residual stresses, the cutting speed should be increased, the uncut chip thickness (or feed) should be reduced and machining with honed tools having large cutting edge radii produce better results than chamfered tools. Moreover, increasing tool wear increases the magnitude of surface residual stresses.

  17. A Review of Current Machine Learning Methods Used for Cancer Recurrence Modeling and Prediction

    SciTech Connect

    Hemphill, Geralyn M.

    Cancer has been characterized as a heterogeneous disease consisting of many different subtypes. The early diagnosis and prognosis of a cancer type has become a necessity in cancer research. A major challenge in cancer management is the classification of patients into appropriate risk groups for better treatment and follow-up. Such risk assessment is critically important in order to optimize the patient’s health and the use of medical resources, as well as to avoid cancer recurrence. This paper focuses on the application of machine learning methods for predicting the likelihood of a recurrence of cancer. It is not meant to bemore » an extensive review of the literature on the subject of machine learning techniques for cancer recurrence modeling. Other recent papers have performed such a review, and I will rely heavily on the results and outcomes from these papers. The electronic databases that were used for this review include PubMed, Google, and Google Scholar. Query terms used include “cancer recurrence modeling”, “cancer recurrence and machine learning”, “cancer recurrence modeling and machine learning”, and “machine learning for cancer recurrence and prediction”. The most recent and most applicable papers to the topic of this review have been included in the references. It also includes a list of modeling and classification methods to predict cancer recurrence.« less

  18. Eigenvalue Solvers for Modeling Nuclear Reactors on Leadership Class Machines

    SciTech Connect

    Slaybaugh, R. N.; Ramirez-Zweiger, M.; Pandya, Tara

    In this paper, three complementary methods have been implemented in the code Denovo that accelerate neutral particle transport calculations with methods that use leadership-class computers fully and effectively: a multigroup block (MG) Krylov solver, a Rayleigh quotient iteration (RQI) eigenvalue solver, and a multigrid in energy (MGE) preconditioner. The MG Krylov solver converges more quickly than Gauss Seidel and enables energy decomposition such that Denovo can scale to hundreds of thousands of cores. RQI should converge in fewer iterations than power iteration (PI) for large and challenging problems. RQI creates shifted systems that would not be tractable without the MGmore » Krylov solver. It also creates ill-conditioned matrices. The MGE preconditioner reduces iteration count significantly when used with RQI and takes advantage of the new energy decomposition such that it can scale efficiently. Each individual method has been described before, but this is the first time they have been demonstrated to work together effectively. The combination of solvers enables the RQI eigenvalue solver to work better than the other available solvers for large reactors problems on leadership-class machines. Using these methods together, RQI converged in fewer iterations and in less time than PI for a full pressurized water reactor core. These solvers also performed better than an Arnoldi eigenvalue solver for a reactor benchmark problem when energy decomposition is needed. The MG Krylov, MGE preconditioner, and RQI solver combination also scales well in energy. Finally, this solver set is a strong choice for very large and challenging problems.« less

  19. Eigenvalue Solvers for Modeling Nuclear Reactors on Leadership Class Machines

    DOE PAGES

    Slaybaugh, R. N.; Ramirez-Zweiger, M.; Pandya, Tara; ...

    2018-02-20

    In this paper, three complementary methods have been implemented in the code Denovo that accelerate neutral particle transport calculations with methods that use leadership-class computers fully and effectively: a multigroup block (MG) Krylov solver, a Rayleigh quotient iteration (RQI) eigenvalue solver, and a multigrid in energy (MGE) preconditioner. The MG Krylov solver converges more quickly than Gauss Seidel and enables energy decomposition such that Denovo can scale to hundreds of thousands of cores. RQI should converge in fewer iterations than power iteration (PI) for large and challenging problems. RQI creates shifted systems that would not be tractable without the MGmore » Krylov solver. It also creates ill-conditioned matrices. The MGE preconditioner reduces iteration count significantly when used with RQI and takes advantage of the new energy decomposition such that it can scale efficiently. Each individual method has been described before, but this is the first time they have been demonstrated to work together effectively. The combination of solvers enables the RQI eigenvalue solver to work better than the other available solvers for large reactors problems on leadership-class machines. Using these methods together, RQI converged in fewer iterations and in less time than PI for a full pressurized water reactor core. These solvers also performed better than an Arnoldi eigenvalue solver for a reactor benchmark problem when energy decomposition is needed. The MG Krylov, MGE preconditioner, and RQI solver combination also scales well in energy. Finally, this solver set is a strong choice for very large and challenging problems.« less

  20. Machine learning for many-body physics: The case of the Anderson impurity model

    SciTech Connect

    Arsenault, Louis-François; Lopez-Bezanilla, Alejandro; von Lilienfeld, O. Anatole

    We applied machine learning methods in order to find the Green's function of the Anderson impurity model, a basic model system of quantum many-body condensed-matter physics. Furthermore, different methods of parametrizing the Green's function are investigated; a representation in terms of Legendre polynomials is found to be superior due to its limited number of coefficients and its applicability to state of the art methods of solution. The dependence of the errors on the size of the training set is determined. Our results indicate that a machine learning approach to dynamical mean-field theory may be feasible.

  1. Machine learning for many-body physics: The case of the Anderson impurity model

    DOE PAGES

    Arsenault, Louis-François; Lopez-Bezanilla, Alejandro; von Lilienfeld, O. Anatole; ...

    2014-10-31

    We applied machine learning methods in order to find the Green's function of the Anderson impurity model, a basic model system of quantum many-body condensed-matter physics. Furthermore, different methods of parametrizing the Green's function are investigated; a representation in terms of Legendre polynomials is found to be superior due to its limited number of coefficients and its applicability to state of the art methods of solution. The dependence of the errors on the size of the training set is determined. Our results indicate that a machine learning approach to dynamical mean-field theory may be feasible.

  2. Nonlinear and Digital Man-machine Control Systems Modeling

    NASA Technical Reports Server (NTRS)

    Mekel, R.

    1972-01-01

    An adaptive modeling technique is examined by which controllers can be synthesized to provide corrective dynamics to a human operator's mathematical model in closed loop control systems. The technique utilizes a class of Liapunov functions formulated for this purpose, Liapunov's stability criterion and a model-reference system configuration. The Liapunov function is formulated to posses variable characteristics to take into consideration the identification dynamics. The time derivative of the Liapunov function generate the identification and control laws for the mathematical model system. These laws permit the realization of a controller which updates the human operator's mathematical model parameters so that model and human operator produce the same response when subjected to the same stimulus. A very useful feature is the development of a digital computer program which is easily implemented and modified concurrent with experimentation. The program permits the modeling process to interact with the experimentation process in a mutually beneficial way.

  3. Teaching Subtraction and Multiplication with Regrouping Using the Concrete-Representational-Abstract Sequence and Strategic Instruction Model

    ERIC Educational Resources Information Center

    Flores, Margaret M.; Hinton, Vanessa; Strozier, Shaunita D.

    2014-01-01

    Based on Common Core Standards (2010), mathematics interventions should emphasize conceptual understanding of numbers and operations as well as fluency. For students at risk for failure, the concrete-representational-abstract (CRA) sequence and the Strategic Instruction Model (SIM) have been shown effective in teaching computation with an emphasis…

  4. Learning with Technology: Video Modeling with Concrete-Representational-Abstract Sequencing for Students with Autism Spectrum Disorder

    ERIC Educational Resources Information Center

    Yakubova, Gulnoza; Hughes, Elizabeth M.; Shinaberry, Megan

    2016-01-01

    The purpose of this study was to determine the effectiveness of a video modeling intervention with concrete-representational-abstract instructional sequence in teaching mathematics concepts to students with autism spectrum disorder (ASD). A multiple baseline across skills design of single-case experimental methodology was used to determine the…

  5. Predicting Mouse Liver Microsomal Stability with “Pruned” Machine Learning Models and Public Data

    PubMed Central

    Perryman, Alexander L.; Stratton, Thomas P.; Ekins, Sean; Freundlich, Joel S.

    2015-01-01

    Purpose Mouse efficacy studies are a critical hurdle to advance translational research of potential therapeutic compounds for many diseases. Although mouse liver microsomal (MLM) stability studies are not a perfect surrogate for in vivo studies of metabolic clearance, they are the initial model system used to assess metabolic stability. Consequently, we explored the development of machine learning models that can enhance the probability of identifying compounds possessing MLM stability. Methods Published assays on MLM half-life values were identified in PubChem, reformatted, and curated to create a training set with 894 unique small molecules. These data were used to construct machine learning models assessed with internal cross-validation, external tests with a published set of antitubercular compounds, and independent validation with an additional diverse set of 571 compounds (PubChem data on percent metabolism). Results “Pruning” out the moderately unstable/moderately stable compounds from the training set produced models with superior predictive power. Bayesian models displayed the best predictive power for identifying compounds with a half-life ≥1 hour. Conclusions Our results suggest the pruning strategy may be of general benefit to improve test set enrichment and provide machine learning models with enhanced predictive value for the MLM stability of small organic molecules. This study represents the most exhaustive study to date of using machine learning approaches with MLM data from public sources. PMID:26415647

  6. Predicting Mouse Liver Microsomal Stability with "Pruned" Machine Learning Models and Public Data.

    PubMed

    Perryman, Alexander L; Stratton, Thomas P; Ekins, Sean; Freundlich, Joel S

    2016-02-01

    Mouse efficacy studies are a critical hurdle to advance translational research of potential therapeutic compounds for many diseases. Although mouse liver microsomal (MLM) stability studies are not a perfect surrogate for in vivo studies of metabolic clearance, they are the initial model system used to assess metabolic stability. Consequently, we explored the development of machine learning models that can enhance the probability of identifying compounds possessing MLM stability. Published assays on MLM half-life values were identified in PubChem, reformatted, and curated to create a training set with 894 unique small molecules. These data were used to construct machine learning models assessed with internal cross-validation, external tests with a published set of antitubercular compounds, and independent validation with an additional diverse set of 571 compounds (PubChem data on percent metabolism). "Pruning" out the moderately unstable / moderately stable compounds from the training set produced models with superior predictive power. Bayesian models displayed the best predictive power for identifying compounds with a half-life ≥1 h. Our results suggest the pruning strategy may be of general benefit to improve test set enrichment and provide machine learning models with enhanced predictive value for the MLM stability of small organic molecules. This study represents the most exhaustive study to date of using machine learning approaches with MLM data from public sources.

  7. Gradient boosting machine for modeling the energy consumption of commercial buildings

    DOE PAGES

    Touzani, Samir; Granderson, Jessica; Fernandes, Samuel

    2017-11-26

    Accurate savings estimations are important to promote energy efficiency projects and demonstrate their cost-effectiveness. The increasing presence of advanced metering infrastructure (AMI) in commercial buildings has resulted in a rising availability of high frequency interval data. These data can be used for a variety of energy efficiency applications such as demand response, fault detection and diagnosis, and heating, ventilation, and air conditioning (HVAC) optimization. This large amount of data has also opened the door to the use of advanced statistical learning models, which hold promise for providing accurate building baseline energy consumption predictions, and thus accurate saving estimations. The gradientmore » boosting machine is a powerful machine learning algorithm that is gaining considerable traction in a wide range of data driven applications, such as ecology, computer vision, and biology. In the present work an energy consumption baseline modeling method based on a gradient boosting machine was proposed. To assess the performance of this method, a recently published testing procedure was used on a large dataset of 410 commercial buildings. The model training periods were varied and several prediction accuracy metrics were used to evaluate the model's performance. The results show that using the gradient boosting machine model improved the R-squared prediction accuracy and the CV(RMSE) in more than 80 percent of the cases, when compared to an industry best practice model that is based on piecewise linear regression, and to a random forest algorithm.« less

  8. Gradient boosting machine for modeling the energy consumption of commercial buildings

    SciTech Connect

    Touzani, Samir; Granderson, Jessica; Fernandes, Samuel

    Accurate savings estimations are important to promote energy efficiency projects and demonstrate their cost-effectiveness. The increasing presence of advanced metering infrastructure (AMI) in commercial buildings has resulted in a rising availability of high frequency interval data. These data can be used for a variety of energy efficiency applications such as demand response, fault detection and diagnosis, and heating, ventilation, and air conditioning (HVAC) optimization. This large amount of data has also opened the door to the use of advanced statistical learning models, which hold promise for providing accurate building baseline energy consumption predictions, and thus accurate saving estimations. The gradientmore » boosting machine is a powerful machine learning algorithm that is gaining considerable traction in a wide range of data driven applications, such as ecology, computer vision, and biology. In the present work an energy consumption baseline modeling method based on a gradient boosting machine was proposed. To assess the performance of this method, a recently published testing procedure was used on a large dataset of 410 commercial buildings. The model training periods were varied and several prediction accuracy metrics were used to evaluate the model's performance. The results show that using the gradient boosting machine model improved the R-squared prediction accuracy and the CV(RMSE) in more than 80 percent of the cases, when compared to an industry best practice model that is based on piecewise linear regression, and to a random forest algorithm.« less

  9. Improving wave forecasting by integrating ensemble modelling and machine learning

    NASA Astrophysics Data System (ADS)

    O'Donncha, F.; Zhang, Y.; James, S. C.

    2017-12-01

    Modern smart-grid networks use technologies to instantly relay information on supply and demand to support effective decision making. Integration of renewable-energy resources with these systems demands accurate forecasting of energy production (and demand) capacities. For wave-energy converters, this requires wave-condition forecasting to enable estimates of energy production. Current operational wave forecasting systems exhibit substantial errors with wave-height RMSEs of 40 to 60 cm being typical, which limits the reliability of energy-generation predictions thereby impeding integration with the distribution grid. In this study, we integrate physics-based models with statistical learning aggregation techniques that combine forecasts from multiple, independent models into a single "best-estimate" prediction of the true state. The Simulating Waves Nearshore physics-based model is used to compute wind- and currents-augmented waves in the Monterey Bay area. Ensembles are developed based on multiple simulations perturbing input data (wave characteristics supplied at the model boundaries and winds) to the model. A learning-aggregation technique uses past observations and past model forecasts to calculate a weight for each model. The aggregated forecasts are compared to observation data to quantify the performance of the model ensemble and aggregation techniques. The appropriately weighted ensemble model outperforms an individual ensemble member with regard to forecasting wave conditions.

  10. A Novel Application of Machine Learning Methods to Model Microcontroller Upset Due to Intentional Electromagnetic Interference

    NASA Astrophysics Data System (ADS)

    Bilalic, Rusmir

    A novel application of support vector machines (SVMs), artificial neural networks (ANNs), and Gaussian processes (GPs) for machine learning (GPML) to model microcontroller unit (MCU) upset due to intentional electromagnetic interference (IEMI) is presented. In this approach, an MCU performs a counting operation (0-7) while electromagnetic interference in the form of a radio frequency (RF) pulse is direct-injected into the MCU clock line. Injection times with respect to the clock signal are the clock low, clock rising edge, clock high, and the clock falling edge periods in the clock window during which the MCU is performing initialization and executing the counting procedure. The intent is to cause disruption in the counting operation and model the probability of effect (PoE) using machine learning tools. Five experiments were executed as part of this research, each of which contained a set of 38,300 training points and 38,300 test points, for a total of 383,000 total points with the following experiment variables: injection times with respect to the clock signal, injected RF power, injected RF pulse width, and injected RF frequency. For the 191,500 training points, the average training error was 12.47%, while for the 191,500 test points the average test error was 14.85%, meaning that on average, the machine was able to predict MCU upset with an 85.15% accuracy. Leaving out the results for the worst-performing model (SVM with a linear kernel), the test prediction accuracy for the remaining machines is almost 89%. All three machine learning methods (ANNs, SVMs, and GPML) showed excellent and consistent results in their ability to model and predict the PoE on an MCU due to IEMI. The GP approach performed best during training with a 7.43% average training error, while the ANN technique was most accurate during the test with a 10.80% error.

  11. Abstraction of the Relational Model from a Department of Veterans Affairs DHCP Database: Bridging Theory and Working Application

    PubMed Central

    Levy, C.; Beauchamp, C.

    1996-01-01

    This poster describes the methods used and working prototype that was developed from an abstraction of the relational model from the VA's hierarchical DHCP database. Overlaying the relational model on DHCP permits multiple user views of the physical data structure, enhances access to the database by providing a link to commercial (SQL based) software, and supports a conceptual managed care data model based on primary and longitudinal patient care. The goal of this work was to create a relational abstraction of the existing hierarchical database; to construct, using SQL data definition language, user views of the database which reflect the clinical conceptual view of DHCP, and to allow the user to work directly with the logical view of the data using GUI based commercial software of their choosing. The workstation is intended to serve as a platform from which a managed care information model could be implemented and evaluated.

  12. State Machine Modeling of the Space Launch System Solid Rocket Boosters

    NASA Technical Reports Server (NTRS)

    Harris, Joshua A.; Patterson-Hine, Ann

    2013-01-01

    The Space Launch System is a Shuttle-derived heavy-lift vehicle currently in development to serve as NASA's premiere launch vehicle for space exploration. The Space Launch System is a multistage rocket with two Solid Rocket Boosters and multiple payloads, including the Multi-Purpose Crew Vehicle. Planned Space Launch System destinations include near-Earth asteroids, the Moon, Mars, and Lagrange points. The Space Launch System is a complex system with many subsystems, requiring considerable systems engineering and integration. To this end, state machine analysis offers a method to support engineering and operational e orts, identify and avert undesirable or potentially hazardous system states, and evaluate system requirements. Finite State Machines model a system as a finite number of states, with transitions between states controlled by state-based and event-based logic. State machines are a useful tool for understanding complex system behaviors and evaluating "what-if" scenarios. This work contributes to a state machine model of the Space Launch System developed at NASA Ames Research Center. The Space Launch System Solid Rocket Booster avionics and ignition subsystems are modeled using MATLAB/Stateflow software. This model is integrated into a larger model of Space Launch System avionics used for verification and validation of Space Launch System operating procedures and design requirements. This includes testing both nominal and o -nominal system states and command sequences.

  13. Publication Abstracts.

    ERIC Educational Resources Information Center

    Johns Hopkins Univ., Baltimore, MD. Center for the Study of Social Organization of Schools.

    This booklet contains abstracts of 62 documents published by the Johns Hopkins University Center for the Study of Social Organization of Schools from September 1967 to May 1970. The majority of the documents are research studies in the areas of desegregation, language development, educational opportunity, and educational games--most of them…

  14. Abstract Constructions.

    ERIC Educational Resources Information Center

    Pietropola, Anne

    1998-01-01

    Describes a lesson designed to culminate a year of eighth-grade art classes in which students explore elements of design and space by creating 3-D abstract constructions. Outlines the process of using foam board and markers to create various shapes and optical effects. (DSK)

  15. Scientist-Centered Workflow Abstractions via Generic Actors, Workflow Templates, and Context-Awareness for Groundwater Modeling and Analysis

    SciTech Connect

    Chin, George; Sivaramakrishnan, Chandrika; Critchlow, Terence J.

    2011-07-04

    A drawback of existing scientific workflow systems is the lack of support to domain scientists in designing and executing their own scientific workflows. Many domain scientists avoid developing and using workflows because the basic objects of workflows are too low-level and high-level tools and mechanisms to aid in workflow construction and use are largely unavailable. In our research, we are prototyping higher-level abstractions and tools to better support scientists in their workflow activities. Specifically, we are developing generic actors that provide abstract interfaces to specific functionality, workflow templates that encapsulate workflow and data patterns that can be reused and adaptedmore » by scientists, and context-awareness mechanisms to gather contextual information from the workflow environment on behalf of the scientist. To evaluate these scientist-centered abstractions on real problems, we apply them to construct and execute scientific workflows in the specific domain area of groundwater modeling and analysis.« less

  16. Comparing and Validating Machine Learning Models for Mycobacterium tuberculosis Drug Discovery.

    PubMed

    Lane, Thomas; Russo, Daniel P; Zorn, Kimberley M; Clark, Alex M; Korotcov, Alexandru; Tkachenko, Valery; Reynolds, Robert C; Perryman, Alexander L; Freundlich, Joel S; Ekins, Sean

    2018-04-26

    Tuberculosis is a global health dilemma. In 2016, the WHO reported 10.4 million incidences and 1.7 million deaths. The need to develop new treatments for those infected with Mycobacterium tuberculosis ( Mtb) has led to many large-scale phenotypic screens and many thousands of new active compounds identified in vitro. However, with limited funding, efforts to discover new active molecules against Mtb needs to be more efficient. Several computational machine learning approaches have been shown to have good enrichment and hit rates. We have curated small molecule Mtb data and developed new models with a total of 18,886 molecules with activity cutoffs of 10 μM, 1 μM, and 100 nM. These data sets were used to evaluate different machine learning methods (including deep learning) and metrics and to generate predictions for additional molecules published in 2017. One Mtb model, a combined in vitro and in vivo data Bayesian model at a 100 nM activity yielded the following metrics for 5-fold cross validation: accuracy = 0.88, precision = 0.22, recall = 0.91, specificity = 0.88, kappa = 0.31, and MCC = 0.41. We have also curated an evaluation set ( n = 153 compounds) published in 2017, and when used to test our model, it showed the comparable statistics (accuracy = 0.83, precision = 0.27, recall = 1.00, specificity = 0.81, kappa = 0.36, and MCC = 0.47). We have also compared these models with additional machine learning algorithms showing Bayesian machine learning models constructed with literature Mtb data generated by different laboratories generally were equivalent to or outperformed deep neural networks with external test sets. Finally, we have also compared our training and test sets to show they were suitably diverse and different in order to represent useful evaluation sets. Such Mtb machine learning models could help prioritize compounds for testing in vitro and in vivo.

  17. Theoretical models of the electrical discharge machining process. III. The variable mass, cylindrical plasma model

    NASA Astrophysics Data System (ADS)

    Eubank, Philip T.; Patel, Mukund R.; Barrufet, Maria A.; Bozkurt, Bedri

    1993-06-01

    A variable mass, cylindrical plasma model (VMCPM) is developed for sparks created by electrical discharge in a liquid media. The model consist of three differential equations—one each from fluid dynamics, an energy balance, and the radiation equation—combined with a plasma equation of state. A thermophysical property subroutine allows realistic estimation of plasma enthalpy, mass density, and particle fractions by inclusion of the heats of dissociation and ionization for a plasma created from deionized water. Problems with the zero-time boundary conditions are overcome by an electron balance procedure. Numerical solution of the model provides plasma radius, temperature, pressure, and mass as a function of pulse time for fixed current, electrode gap, and power fraction remaining in the plasma. Moderately high temperatures (≳5000 K) and pressures (≳4 bar) persist in the sparks even after long pulse times (to ˜500 μs). Quantitative proof that superheating is the dominant mechanism for electrical discharge machining (EDM) erosion is thus provided for the first time. Some quantitative inconsistencies developed between our (1) cathode, (2) anode, and (3) plasma models (this series) are discussed with indication as to how they will be rectified in a fourth article to follow shortly in this journal. While containing oversimplifications, these three models are believed to contain the respective dominant physics of the EDM process but need be brought into numerical consistency for each time increment of the numerical solution.

  18. SAINT: A combined simulation language for modeling man-machine systems

    NASA Technical Reports Server (NTRS)

    Seifert, D. J.

    1979-01-01

    SAINT (Systems Analysis of Integrated Networks of Tasks) is a network modeling and simulation technique for design and analysis of complex man machine systems. SAINT provides the conceptual framework for representing systems that consist of discrete task elements, continuous state variables, and interactions between them. It also provides a mechanism for combining human performance models and dynamic system behaviors in a single modeling structure. The SAINT technique is described and applications of the SAINT are discussed.

  19. Mapping groundwater contamination risk of multiple aquifers using multi-model ensemble of machine learning algorithms.

    PubMed

    Barzegar, Rahim; Moghaddam, Asghar Asghari; Deo, Ravinesh; Fijani, Elham; Tziritis, Evangelos

    2018-04-15

    Constructing accurate and reliable groundwater risk maps provide scientifically prudent and strategic measures for the protection and management of groundwater. The objectives of this paper are to design and validate machine learning based-risk maps using ensemble-based modelling with an integrative approach. We employ the extreme learning machines (ELM), multivariate regression splines (MARS), M5 Tree and support vector regression (SVR) applied in multiple aquifer systems (e.g. unconfined, semi-confined and confined) in the Marand plain, North West Iran, to encapsulate the merits of individual learning algorithms in a final committee-based ANN model. The DRASTIC Vulnerability Index (VI) ranged from 56.7 to 128.1, categorized with no risk, low and moderate vulnerability thresholds. The correlation coefficient (r) and Willmott's Index (d) between NO 3 concentrations and VI were 0.64 and 0.314, respectively. To introduce improvements in the original DRASTIC method, the vulnerability indices were adjusted by NO 3 concentrations, termed as the groundwater contamination risk (GCR). Seven DRASTIC parameters utilized as the model inputs and GCR values utilized as the outputs of individual machine learning models were served in the fully optimized committee-based ANN-predictive model. The correlation indicators demonstrated that the ELM and SVR models outperformed the MARS and M5 Tree models, by virtue of a larger d and r value. Subsequently, the r and d metrics for the ANN-committee based multi-model in the testing phase were 0.8889 and 0.7913, respectively; revealing the superiority of the integrated (or ensemble) machine learning models when compared with the original DRASTIC approach. The newly designed multi-model ensemble-based approach can be considered as a pragmatic step for mapping groundwater contamination risks of multiple aquifer systems with multi-model techniques, yielding the high accuracy of the ANN committee-based model. Copyright © 2017 Elsevier B

  20. Analysis of precision and accuracy in a simple model of machine learning

    NASA Astrophysics Data System (ADS)

    Lee, Julian

    2017-12-01

    Machine learning is a procedure where a model for the world is constructed from a training set of examples. It is important that the model should capture relevant features of the training set, and at the same time make correct prediction for examples not included in the training set. I consider the polynomial regression, the simplest method of learning, and analyze the accuracy and precision for different levels of the model complexity.

  1. Modeling and analysis of dynamic characteristics of carrier system of machining center in MSC.Adams

    NASA Astrophysics Data System (ADS)

    Grinek, A. V.; Rybina, A. V.; Boychuk, I. P.; Dantsevich, I. M.; Hurtasenko, A. V.

    2018-03-01

    The simulation model with the help of vibration analysis was developed in MSC.Adams/Vibration and experimental research of the dynamic characteristics of a five-axis machining center was carried out. The amplitude-frequency characteristics, resonant frequencies in various directions are investigated. Dynamic and static rigidity, damping intensity and the coefficient of dynamism of the center are determined.

  2. Modeling and simulation of the fluid flow in wire electrochemical machining with rotating tool (wire ECM)

    NASA Astrophysics Data System (ADS)

    Klocke, F.; Herrig, T.; Zeis, M.; Klink, A.

    2017-10-01

    Combining the working principle of electrochemical machining (ECM) with a universal rotating tool, like a wire, could manage lots of challenges of the classical ECM sinking process. Such a wire-ECM process could be able to machine flexible and efficient 2.5-dimensional geometries like fir tree slots in turbine discs. Nowadays, established manufacturing technologies for slotting turbine discs are broaching and wire electrical discharge machining (wire EDM). Nevertheless, high requirements on surface integrity of turbine parts need cost intensive process development and - in case of wire-EDM - trim cuts to reduce the heat affected rim zone. Due to the process specific advantages, ECM is an attractive alternative manufacturing technology and is getting more and more relevant for sinking applications within the last few years. But ECM is also opposed with high costs for process development and complex electrolyte flow devices. In the past, few studies dealt with the development of a wire ECM process to meet these challenges. However, previous concepts of wire ECM were only suitable for micro machining applications. Due to insufficient flushing concepts the application of the process for machining macro geometries failed. Therefore, this paper presents the modeling and simulation of a new flushing approach for process assessment. The suitability of a rotating structured wire electrode in combination with an axial flushing for electrodes with high aspect ratios is investigated and discussed.

  3. Modelling rollover behaviour of exacavator-based forest machines

    Treesearch

    M.W. Veal; S.E. Taylor; Robert B. Rummer

    2003-01-01

    This poster presentation provides results from analytical and computer simulation models of rollover behaviour of hydraulic excavators. These results are being used as input to the operator protective structure standards development process. Results from rigid body mechanics and computer simulation methods agree well with field rollover test data. These results show...

  4. Fault Modeling of Extreme Scale Applications Using Machine Learning

    SciTech Connect

    Vishnu, Abhinav; Dam, Hubertus van; Tallent, Nathan R.

    Faults are commonplace in large scale systems. These systems experience a variety of faults such as transient, permanent and intermittent. Multi-bit faults are typically not corrected by the hardware resulting in an error. Here, this paper attempts to answer an important question: Given a multi-bit fault in main memory, will it result in an application error — and hence a recovery algorithm should be invoked — or can it be safely ignored? We propose an application fault modeling methodology to answer this question. Given a fault signature (a set of attributes comprising of system and application state), we use machinemore » learning to create a model which predicts whether a multibit permanent/transient main memory fault will likely result in error. We present the design elements such as the fault injection methodology for covering important data structures, the application and system attributes which should be used for learning the model, the supervised learning algorithms (and potentially ensembles), and important metrics. Lastly, we use three applications — NWChem, LULESH and SVM — as examples for demonstrating the effectiveness of the proposed fault modeling methodology.« less

  5. Fault Modeling of Extreme Scale Applications Using Machine Learning

    DOE PAGES

    Vishnu, Abhinav; Dam, Hubertus van; Tallent, Nathan R.; ...

    2016-05-01

    Faults are commonplace in large scale systems. These systems experience a variety of faults such as transient, permanent and intermittent. Multi-bit faults are typically not corrected by the hardware resulting in an error. Here, this paper attempts to answer an important question: Given a multi-bit fault in main memory, will it result in an application error — and hence a recovery algorithm should be invoked — or can it be safely ignored? We propose an application fault modeling methodology to answer this question. Given a fault signature (a set of attributes comprising of system and application state), we use machinemore » learning to create a model which predicts whether a multibit permanent/transient main memory fault will likely result in error. We present the design elements such as the fault injection methodology for covering important data structures, the application and system attributes which should be used for learning the model, the supervised learning algorithms (and potentially ensembles), and important metrics. Lastly, we use three applications — NWChem, LULESH and SVM — as examples for demonstrating the effectiveness of the proposed fault modeling methodology.« less

  6. Hamiltonian methods of modeling and control of AC microgrids with spinning machines and inverters

    SciTech Connect

    Matthews, Ronald C.; Weaver, Wayne W.; Robinett, Rush D.

    This study presents a novel approach to the modeling and control of AC microgrids that contain spinning machines, power electronic inverters and energy storage devices. The inverters in the system can adjust their frequencies and power angles very quickly, so the modeling focuses on establishing a common reference frequency and angle in the microgrid based on the spinning machines. From this dynamic model, nonlinear Hamiltonian surface shaping and power flow control method is applied and shown to stabilize. From this approach the energy flow in the system is used to show the energy storage device requirements and limitations for themore » system. This paper first describes the model for a single bus AC microgrid with a Hamiltonian control, then extends this model and control to a more general class of multiple bus AC microgrids. Finally, simulation results demonstrate the efficacy of the approach in stabilizing and optimization of the microgrid.« less

  7. Hamiltonian methods of modeling and control of AC microgrids with spinning machines and inverters

    DOE PAGES

    Matthews, Ronald C.; Weaver, Wayne W.; Robinett, Rush D.; ...

    2017-12-22

    This study presents a novel approach to the modeling and control of AC microgrids that contain spinning machines, power electronic inverters and energy storage devices. The inverters in the system can adjust their frequencies and power angles very quickly, so the modeling focuses on establishing a common reference frequency and angle in the microgrid based on the spinning machines. From this dynamic model, nonlinear Hamiltonian surface shaping and power flow control method is applied and shown to stabilize. From this approach the energy flow in the system is used to show the energy storage device requirements and limitations for themore » system. This paper first describes the model for a single bus AC microgrid with a Hamiltonian control, then extends this model and control to a more general class of multiple bus AC microgrids. Finally, simulation results demonstrate the efficacy of the approach in stabilizing and optimization of the microgrid.« less

  8. Programming and machining of complex parts based on CATIA solid modeling

    NASA Astrophysics Data System (ADS)

    Zhu, Xiurong

    2017-09-01

    The complex parts of the use of CATIA solid modeling programming and simulation processing design, elaborated in the field of CNC machining, programming and the importance of processing technology. In parts of the design process, first make a deep analysis on the principle, and then the size of the design, the size of each chain, connected to each other. After the use of backstepping and a variety of methods to calculate the final size of the parts. In the selection of parts materials, careful study, repeated testing, the final choice of 6061 aluminum alloy. According to the actual situation of the processing site, it is necessary to make a comprehensive consideration of various factors in the machining process. The simulation process should be based on the actual processing, not only pay attention to shape. It can be used as reference for machining.

  9. A prognostic model for temporal courses that combines temporal abstraction and case-based reasoning.

    PubMed

    Schmidt, Rainer; Gierl, Lothar

    2005-03-01

    Since clinical management of patients and clinical research are essentially time-oriented endeavours, reasoning about time has become a hot topic in medical informatics. Here we present a method for prognosis of temporal courses, which combines temporal abstractions with case-based reasoning. It is useful for application domains where neither well-known standards, nor known periodicity, nor a complete domain theory exist. We have used our method in two prognostic applications. The first one deals with prognosis of the kidney function for intensive care patients. The idea is to elicit impairments on time, especially to warn against threatening kidney failures. Our second application deals with a completely different domain, namely geographical medicine. Its intention is to compute early warnings against approaching infectious diseases, which are characterised by irregular cyclic occurrences. So far, we have applied our program on influenza and bronchitis. In this paper, we focus on influenza forecast and show first experimental results.

  10. Shot-by-shot Spectrum Model for Rod-pinch, Pulsed Radiography Machines

    SciTech Connect

    Wood, William Monford

    A simplified model of bremsstrahlung production is developed for determining the x-ray spectrum output of a rod-pinch radiography machine, on a shot-by-shot basis, using the measured voltage, V(t), and current, I(t). The motivation for this model is the need for an agile means of providing shot-by-shot spectrum prediction, from a laptop or desktop computer, for quantitative radiographic analysis. Simplifying assumptions are discussed, and the model is applied to the Cygnus rod-pinch machine. Output is compared to wedge transmission data for a series of radiographs from shots with identical target objects. Resulting model enables variation of parameters in real time, thusmore » allowing for rapid optimization of the model across many shots. “Goodness of fit” is compared with output from LSP Particle-In-Cell code, as well as the Monte Carlo Neutron Propagation with Xrays (“MCNPX”) model codes, and is shown to provide an excellent predictive representation of the spectral output of the Cygnus machine. In conclusion, improvements to the model, specifically for application to other geometries, are discussed.« less

  11. Shot-by-shot Spectrum Model for Rod-pinch, Pulsed Radiography Machines

    DOE PAGES

    Wood, William Monford

    2018-02-07

    A simplified model of bremsstrahlung production is developed for determining the x-ray spectrum output of a rod-pinch radiography machine, on a shot-by-shot basis, using the measured voltage, V(t), and current, I(t). The motivation for this model is the need for an agile means of providing shot-by-shot spectrum prediction, from a laptop or desktop computer, for quantitative radiographic analysis. Simplifying assumptions are discussed, and the model is applied to the Cygnus rod-pinch machine. Output is compared to wedge transmission data for a series of radiographs from shots with identical target objects. Resulting model enables variation of parameters in real time, thusmore » allowing for rapid optimization of the model across many shots. “Goodness of fit” is compared with output from LSP Particle-In-Cell code, as well as the Monte Carlo Neutron Propagation with Xrays (“MCNPX”) model codes, and is shown to provide an excellent predictive representation of the spectral output of the Cygnus machine. In conclusion, improvements to the model, specifically for application to other geometries, are discussed.« less

  12. Modeling of the flow stress for AISI H13 Tool Steel during Hard Machining Processes

    NASA Astrophysics Data System (ADS)

    Umbrello, Domenico; Rizzuti, Stefania; Outeiro, José C.; Shivpuri, Rajiv

    2007-04-01

    In general, the flow stress models used in computer simulation of machining processes are a function of effective strain, effective strain rate and temperature developed during the cutting process. However, these models do not adequately describe the material behavior in hard machining, where a range of material hardness between 45 and 60 HRC are used. Thus, depending on the specific material hardness different material models must be used in modeling the cutting process. This paper describes the development of a hardness-based flow stress and fracture models for the AISI H13 tool steel, which can be applied for range of material hardness mentioned above. These models were implemented in a non-isothermal viscoplastic numerical model to simulate the machining process for AISI H13 with various hardness values and applying different cutting regime parameters. Predicted results are validated by comparing them with experimental results found in the literature. They are found to predict reasonably well the cutting forces as well as the change in chip morphology from continuous to segmented chip as the material hardness change.

  13. A tool for urban soundscape evaluation applying Support Vector Machines for developing a soundscape classification model.

    PubMed

    Torija, Antonio J; Ruiz, Diego P; Ramos-Ridao, Angel F

    2014-06-01

    To ensure appropriate soundscape management in urban environments, the urban-planning authorities need a range of tools that enable such a task to be performed. An essential step during the management of urban areas from a sound standpoint should be the evaluation of the soundscape in such an area. In this sense, it has been widely acknowledged that a subjective and acoustical categorization of a soundscape is the first step to evaluate it, providing a basis for designing or adapting it to match people's expectations as well. In this sense, this work proposes a model for automatic classification of urban soundscapes. This model is intended for the automatic classification of urban soundscapes based on underlying acoustical and perceptual criteria. Thus, this classification model is proposed to be used as a tool for a comprehensive urban soundscape evaluation. Because of the great complexity associated with the problem, two machine learning techniques, Support Vector Machines (SVM) and Support Vector Machines trained with Sequential Minimal Optimization (SMO), are implemented in developing model classification. The results indicate that the SMO model outperforms the SVM model in the specific task of soundscape classification. With the implementation of the SMO algorithm, the classification model achieves an outstanding performance (91.3% of instances correctly classified). © 2013 Elsevier B.V. All rights reserved.

  14. Shot-by-shot spectrum model for rod-pinch, pulsed radiography machines

    NASA Astrophysics Data System (ADS)

    Wood, Wm M.

    2018-02-01

    A simplified model of bremsstrahlung production is developed for determining the x-ray spectrum output of a rod-pinch radiography machine, on a shot-by-shot basis, using the measured voltage, V(t), and current, I(t). The motivation for this model is the need for an agile means of providing shot-by-shot spectrum prediction, from a laptop or desktop computer, for quantitative radiographic analysis. Simplifying assumptions are discussed, and the model is applied to the Cygnus rod-pinch machine. Output is compared to wedge transmission data for a series of radiographs from shots with identical target objects. Resulting model enables variation of parameters in real time, thus allowing for rapid optimization of the model across many shots. "Goodness of fit" is compared with output from LSP Particle-In-Cell code, as well as the Monte Carlo Neutron Propagation with Xrays ("MCNPX") model codes, and is shown to provide an excellent predictive representation of the spectral output of the Cygnus machine. Improvements to the model, specifically for application to other geometries, are discussed.

  15. Limitations Of The Current State Space Modelling Approach In Multistage Machining Processes Due To Operation Variations

    NASA Astrophysics Data System (ADS)

    Abellán-Nebot, J. V.; Liu, J.; Romero, F.

    2009-11-01

    The State Space modelling approach has been recently proposed as an engineering-driven technique for part quality prediction in Multistage Machining Processes (MMP). Current State Space models incorporate fixture and datum variations in the multi-stage variation propagation, without explicitly considering common operation variations such as machine-tool thermal distortions, cutting-tool wear, cutting-tool deflections, etc. This paper shows the limitations of the current State Space model through an experimental case study where the effect of the spindle thermal expansion, cutting-tool flank wear and locator errors are introduced. The paper also discusses the extension of the current State Space model to include operation variations and its potential benefits.

  16. Data Mining and Machine Learning Models for Predicting Drug Likeness and their Disease or Organ Category

    NASA Astrophysics Data System (ADS)

    Yosipof, Abraham; Guedes, Rita C.; García-Sosa, Alfonso T.

    2018-05-01

    Data mining approaches can uncover underlying patterns in chemical and pharmacological property space decisive for drug discovery and development. Two of the most common approaches are visualization and machine learning methods. Visualization methods use dimensionality reduction techniques in order to reduce multi-dimension data into 2D or 3D representations with a minimal loss of information. Machine learning attempts to find correlations between specific activities or classifications for a set of compounds and their features by means of recurring mathematical models. Both models take advantage of the different and deep relationships that can exist between features of compounds, and helpfully provide classification of compounds based on such features. Drug-likeness has been studied from several viewpoints, but here we provide the first implementation in chemoinformatics of the t-Distributed Stochastic Neighbor Embedding (t-SNE) method for the visualization and the representation of chemical space, and the use of different machine learning methods separately and together to form a new ensemble learning method called AL Boost. The models obtained from AL Boost synergistically combine decision tree, random forests (RF), support vector machine (SVM), artificial neuronal network (ANN), k nearest neighbors (kNN), and logistic regression models. In this work, we show that together they form a predictive model that not only improves the predictive force but also decreases bias. This resulted in a corrected classification rate of over 0.81, as well as higher sensitivity and specificity rates for the models. In addition, separation and good models were also achieved for disease categories such as antineoplastic compounds and nervous system diseases, among others. Such models can be used to guide decision on the feature landscape of compounds and their likeness to either drugs or other characteristics, such as specific or multiple disease-category(ies) or organ(s) of action

  17. Data Mining and Machine Learning Models for Predicting Drug Likeness and Their Disease or Organ Category.

    PubMed

    Yosipof, Abraham; Guedes, Rita C; García-Sosa, Alfonso T

    2018-01-01

    Data mining approaches can uncover underlying patterns in chemical and pharmacological property space decisive for drug discovery and development. Two of the most common approaches are visualization and machine learning methods. Visualization methods use dimensionality reduction techniques in order to reduce multi-dimension data into 2D or 3D representations with a minimal loss of information. Machine learning attempts to find correlations between specific activities or classifications for a set of compounds and their features by means of recurring mathematical models. Both models take advantage of the different and deep relationships that can exist between features of compounds, and helpfully provide classification of compounds based on such features or in case of visualization methods uncover underlying patterns in the feature space. Drug-likeness has been studied from several viewpoints, but here we provide the first implementation in chemoinformatics of the t-Distributed Stochastic Neighbor Embedding (t-SNE) method for the visualization and the representation of chemical space, and the use of different machine learning methods separately and together to form a new ensemble learning method called AL Boost. The models obtained from AL Boost synergistically combine decision tree, random forests (RF), support vector machine (SVM), artificial neural network (ANN), k nearest neighbors (kNN), and logistic regression models. In this work, we show that together they form a predictive model that not only improves the predictive force but also decreases bias. This resulted in a corrected classification rate of over 0.81, as well as higher sensitivity and specificity rates for the models. In addition, separation and good models were also achieved for disease categories such as antineoplastic compounds and nervous system diseases, among others. Such models can be used to guide decision on the feature landscape of compounds and their likeness to either drugs or other

  18. Lateral-Directional Parameter Estimation on the X-48B Aircraft Using an Abstracted, Multi-Objective Effector Model

    NASA Technical Reports Server (NTRS)

    Ratnayake, Nalin A.; Waggoner, Erin R.; Taylor, Brian R.

    2011-01-01

    The problem of parameter estimation on hybrid-wing-body aircraft is complicated by the fact that many design candidates for such aircraft involve a large number of aerodynamic control effectors that act in coplanar motion. This adds to the complexity already present in the parameter estimation problem for any aircraft with a closed-loop control system. Decorrelation of flight and simulation data must be performed in order to ascertain individual surface derivatives with any sort of mathematical confidence. Non-standard control surface configurations, such as clamshell surfaces and drag-rudder modes, further complicate the modeling task. In this paper, time-decorrelation techniques are applied to a model structure selected through stepwise regression for simulated and flight-generated lateral-directional parameter estimation data. A virtual effector model that uses mathematical abstractions to describe the multi-axis effects of clamshell surfaces is developed and applied. Comparisons are made between time history reconstructions and observed data in order to assess the accuracy of the regression model. The Cram r-Rao lower bounds of the estimated parameters are used to assess the uncertainty of the regression model relative to alternative models. Stepwise regression was found to be a useful technique for lateral-directional model design for hybrid-wing-body aircraft, as suggested by available flight data. Based on the results of this study, linear regression parameter estimation methods using abstracted effectors are expected to perform well for hybrid-wing-body aircraft properly equipped for the task.

  19. Analytical Modeling of a Novel Transverse Flux Machine for Direct Drive Wind Turbine Applications: Preprint

    SciTech Connect

    Hasan, IIftekhar; Husain, Tausif; Uddin, Md Wasi

    2015-08-24

    This paper presents a nonlinear analytical model of a novel double-sided flux concentrating Transverse Flux Machine (TFM) based on the Magnetic Equivalent Circuit (MEC) model. The analytical model uses a series-parallel combination of flux tubes to predict the flux paths through different parts of the machine including air gaps, permanent magnets, stator, and rotor. The two-dimensional MEC model approximates the complex three-dimensional flux paths of the TFM and includes the effects of magnetic saturation. The model is capable of adapting to any geometry that makes it a good alternative for evaluating prospective designs of TFM compared to finite element solversmore » that are numerically intensive and require more computation time. A single-phase, 1-kW, 400-rpm machine is analytically modeled, and its resulting flux distribution, no-load EMF, and torque are verified with finite element analysis. The results are found to be in agreement, with less than 5% error, while reducing the computation time by 25 times.« less

  20. Analytical Modeling of a Novel Transverse Flux Machine for Direct Drive Wind Turbine Applications

    SciTech Connect

    Hasan, IIftekhar; Husain, Tausif; Uddin, Md Wasi

    2015-09-02

    This paper presents a nonlinear analytical model of a novel double sided flux concentrating Transverse Flux Machine (TFM) based on the Magnetic Equivalent Circuit (MEC) model. The analytical model uses a series-parallel combination of flux tubes to predict the flux paths through different parts of the machine including air gaps, permanent magnets (PM), stator, and rotor. The two-dimensional MEC model approximates the complex three-dimensional flux paths of the TFM and includes the effects of magnetic saturation. The model is capable of adapting to any geometry which makes it a good alternative for evaluating prospective designs of TFM as compared tomore » finite element solvers which are numerically intensive and require more computation time. A single phase, 1 kW, 400 rpm machine is analytically modeled and its resulting flux distribution, no-load EMF and torque, verified with Finite Element Analysis (FEA). The results are found to be in agreement with less than 5% error, while reducing the computation time by 25 times.« less

  1. Experience with abstract notation one

    NASA Technical Reports Server (NTRS)

    Harvey, James D.; Weaver, Alfred C.

    1990-01-01

    The development of computer science has produced a vast number of machine architectures, programming languages, and compiler technologies. The cross product of these three characteristics defines the spectrum of previous and present data representation methodologies. With regard to computer networks, the uniqueness of these methodologies presents an obstacle when disparate host environments are to be interconnected. Interoperability within a heterogeneous network relies upon the establishment of data representation commonality. The International Standards Organization (ISO) is currently developing the abstract syntax notation one standard (ASN.1) and the basic encoding rules standard (BER) that collectively address this problem. When used within the presentation layer of the open systems interconnection reference model, these two standards provide the data representation commonality required to facilitate interoperability. The details of a compiler that was built to automate the use of ASN.1 and BER are described. From this experience, insights into both standards are given and potential problems relating to this development effort are discussed.

  2. (abstract) Generic Modeling of a Life Support System for Process Technology Comparisons

    NASA Technical Reports Server (NTRS)

    Ferrall, J. F.; Seshan, P. K.; Rohatgi, N. K.; Ganapathi, G. B.

    1993-01-01

    This paper describes a simulation model called the Life Support Systems Analysis Simulation Tool (LiSSA-ST), the spreadsheet program called the Life Support Systems Analysis Trade Tool (LiSSA-TT), and the Generic Modular Flow Schematic (GMFS) modeling technique. Results of using the LiSSA-ST and the LiSSA-TT will be presented for comparing life support systems and process technology options for a Lunar Base and a Mars Exploration Mission.

  3. Sensitivity analysis of machine-learning models of hydrologic time series

    NASA Astrophysics Data System (ADS)

    O'Reilly, A. M.

    2017-12-01

    Sensitivity analysis traditionally has been applied to assessing model response to perturbations in model parameters, where the parameters are those model input variables adjusted during calibration. Unlike physics-based models where parameters represent real phenomena, the equivalent of parameters for machine-learning models are simply mathematical "knobs" that are automatically adjusted during training/testing/verification procedures. Thus the challenge of extracting knowledge of hydrologic system functionality from machine-learning models lies in their very nature, leading to the label "black box." Sensitivity analysis of the forcing-response behavior of machine-learning models, however, can provide understanding of how the physical phenomena represented by model inputs affect the physical phenomena represented by model outputs.As part of a previous study, hybrid spectral-decomposition artificial neural network (ANN) models were developed to simulate the observed behavior of hydrologic response contained in multidecadal datasets of lake water level, groundwater level, and spring flow. Model inputs used moving window averages (MWA) to represent various frequencies and frequency-band components of time series of rainfall and groundwater use. Using these forcing time series, the MWA-ANN models were trained to predict time series of lake water level, groundwater level, and spring flow at 51 sites in central Florida, USA. A time series of sensitivities for each MWA-ANN model was produced by perturbing forcing time-series and computing the change in response time-series per unit change in perturbation. Variations in forcing-response sensitivities are evident between types (lake, groundwater level, or spring), spatially (among sites of the same type), and temporally. Two generally common characteristics among sites are more uniform sensitivities to rainfall over time and notable increases in sensitivities to groundwater usage during significant drought periods.

  4. Thermomechanical Modeling of Sintered Silver - A Fracture Mechanics-based Approach: Extended Abstract: Preprint

    SciTech Connect

    Paret, Paul P; DeVoto, Douglas J; Narumanchi, Sreekant V

    Sintered silver has proven to be a promising candidate for use as a die-attach and substrate-attach material in automotive power electronics components. It holds promise of greater reliability than lead-based and lead-free solders, especially at higher temperatures (less than 200 degrees Celcius). Accurate predictive lifetime models of sintered silver need to be developed and its failure mechanisms thoroughly characterized before it can be deployed as a die-attach or substrate-attach material in wide-bandgap device-based packages. We present a finite element method (FEM) modeling methodology that can offer greater accuracy in predicting the failure of sintered silver under accelerated thermal cycling. Amore » fracture mechanics-based approach is adopted in the FEM model, and J-integral/thermal cycle values are computed. In this paper, we outline the procedures for obtaining the J-integral/thermal cycle values in a computational model and report on the possible advantage of using these values as modeling parameters in a predictive lifetime model.« less

  5. Law machines: scale models, forensic materiality and the making of modern patent law.

    PubMed

    Pottage, Alain

    2011-10-01

    Early US patent law was machine made. Before the Patent Office took on the function of examining patent applications in 1836, questions of novelty and priority were determined in court, within the forum of the infringement action. And at all levels of litigation, from the circuit courts up to the Supreme Court, working models were the media through which doctrine, evidence and argument were made legible, communicated and interpreted. A model could be set on a table, pointed at, picked up, rotated or upended so as to display a point of interest to a particular audience within the courtroom, and, crucially, set in motion to reveal the 'mode of operation' of a machine. The immediate object of demonstration was to distinguish the intangible invention from its tangible embodiment, but models also'machined' patent law itself. Demonstrations of patent claims with models articulated and resolved a set of conceptual tensions that still make the definition and apprehension of the invention difficult, even today, but they resolved these tensions in the register of materiality, performativity and visibility, rather than the register of conceptuality. The story of models tells us something about how inventions emerge and subsist within the context of patent litigation and patent doctrine, and it offers a starting point for renewed reflection on the question of how technology becomes property.

  6. Efficient Embedded Decoding of Neural Network Language Models in a Machine Translation System.

    PubMed

    Zamora-Martinez, Francisco; Castro-Bleda, Maria Jose

    2018-02-22

    Neural Network Language Models (NNLMs) are a successful approach to Natural Language Processing tasks, such as Machine Translation. We introduce in this work a Statistical Machine Translation (SMT) system which fully integrates NNLMs in the decoding stage, breaking the traditional approach based on [Formula: see text]-best list rescoring. The neural net models (both language models (LMs) and translation models) are fully coupled in the decoding stage, allowing to more strongly influence the translation quality. Computational issues were solved by using a novel idea based on memorization and smoothing of the softmax constants to avoid their computation, which introduces a trade-off between LM quality and computational cost. These ideas were studied in a machine translation task with different combinations of neural networks used both as translation models and as target LMs, comparing phrase-based and [Formula: see text]-gram-based systems, showing that the integrated approach seems more promising for [Formula: see text]-gram-based systems, even with nonfull-quality NNLMs.

  7. Algerian Abstract

    NASA Image and Video Library

    2017-12-08

    Algerian Abstract - April 8th, 1985 Description: What look like pale yellow paint streaks slashing through a mosaic of mottled colors are ridges of wind-blown sand that make up Erg Iguidi, an area of ever-shifting sand dunes extending from Algeria into Mauritania in northwestern Africa. Erg Iguidi is one of several Saharan ergs, or sand seas, where individual dunes often surpass 500 meters-nearly a third of a mile-in both width and height. Credit: USGS/NASA/Landsat 5 To learn more about the Landsat satellite go to: landsat.gsfc.nasa.gov/ NASA Goddard Space Flight Center enables NASA’s mission through four scientific endeavors: Earth Science, Heliophysics, Solar System Exploration, and Astrophysics. Goddard plays a leading role in NASA’s accomplishments by contributing compelling scientific knowledge to advance the Agency’s mission. Follow us on Twitter Join us on Facebook

  8. Progress on Fault Mechanisms for Gear Transmissions in Coal Cutting Machines: From Macro to Nano Models.

    PubMed

    Jiang, Yu; Zhang, Xiaogang; Zhang, Chao; Li, Zhixiong; Sheng, Chenxing

    2017-04-01

    Numerical modeling has been recognized as the dispensable tools for mechanical fault mechanism analysis. Techniques, ranging from macro to nano levels, include the finite element modeling boundary element modeling, modular dynamic modeling, nano dynamic modeling and so forth. This work firstly reviewed the progress on the fault mechanism analysis for gear transmissions from the tribological and dynamic aspects. Literature review indicates that the tribological and dynamic properties were separately investigated to explore the fault mechanism in gear transmissions. However, very limited work has been done to address the links between the tribological and dynamic properties and scarce researches have been done for coal cutting machines. For this reason, the tribo-dynamic coupled model was introduced to bridge the gap between the tribological and dynamic models in fault mechanism analysis for gear transmissions in coal cutting machines. The modular dynamic modeling and nano dynamic modeling techniques are expected to establish the links between the tribological and dynamic models. Possible future research directions using the tribo dynamic coupled model were summarized to provide potential references for researchers in the field.

  9. An abstract model for radiative transfer in an atmosphere with reflection by the planetary surface

    NASA Astrophysics Data System (ADS)

    Greenberg, W.; van der Mee, C. V. M.

    1985-07-01

    A Hilbert-space model is developed that applies to radiative transfer in a homogeneous, plane-parallel planetary atmosphere. Reflection and absorption by the planetary surface are taken into account by imposing a reflective boundary condition. The existence and uniqueness of the solution of this boundary value problem are established by proving the invertibility of a scattering operator using the Fredholm alternative.

  10. Using Machine Learning as a fast emulator of physical processes within the Met Office's Unified Model

    NASA Astrophysics Data System (ADS)

    Prudden, R.; Arribas, A.; Tomlinson, J.; Robinson, N.

    2017-12-01

    The Unified Model is a numerical model of the atmosphere used at the UK Met Office (and numerous partner organisations including Korean Meteorological Agency, Australian Bureau of Meteorology and US Air Force) for both weather and climate applications.Especifically, dynamical models such as the Unified Model are now a central part of weather forecasting. Starting from basic physical laws, these models make it possible to predict events such as storms before they have even begun to form. The Unified Model can be simply described as having two components: one component solves the navier-stokes equations (usually referred to as the "dynamics"); the other solves relevant sub-grid physical processes (usually referred to as the "physics"). Running weather forecasts requires substantial computing resources - for example, the UK Met Office operates the largest operational High Performance Computer in Europe - and the cost of a typical simulation is spent roughly 50% in the "dynamics" and 50% in the "physics". Therefore there is a high incentive to reduce cost of weather forecasts and Machine Learning is a possible option because, once a machine learning model has been trained, it is often much faster to run than a full simulation. This is the motivation for a technique called model emulation, the idea being to build a fast statistical model which closely approximates a far more expensive simulation. In this paper we discuss the use of Machine Learning as an emulator to replace the "physics" component of the Unified Model. Various approaches and options will be presented and the implications for further model development, operational running of forecasting systems, development of data assimilation schemes, and development of ensemble prediction techniques will be discussed.

  11. Modeling and predicting abstract concept or idea introduction and propagation through geopolitical groups

    NASA Astrophysics Data System (ADS)

    Jaenisch, Holger M.; Handley, James W.; Hicklen, Michael L.

    2007-04-01

    This paper describes a novel capability for modeling known idea propagation transformations and predicting responses to new ideas from geopolitical groups. Ideas are captured using semantic words that are text based and bear cognitive definitions. We demonstrate a unique algorithm for converting these into analytical predictive equations. Using the illustrative idea of "proposing a gasoline price increase of 1 per gallon from 2" and its changing perceived impact throughout 5 demographic groups, we identify 13 cost of living Diplomatic, Information, Military, and Economic (DIME) features common across all 5 demographic groups. This enables the modeling and monitoring of Political, Military, Economic, Social, Information, and Infrastructure (PMESII) effects of each group to this idea and how their "perception" of this proposal changes. Our algorithm and results are summarized in this paper.

  12. Vehicle Concept Model Abstractions for Integrated Geometric, Inertial, Rigid Body, Powertrain, and FE Analysis

    DTIC Science & Technology

    2011-01-01

    refinement of the vehicle body structure through quantitative assessment of stiffness and modal parameter changes resulting from modifications to the beam...differential placed on the axle , adjustment of the torque output to the opposite wheel may be required to obtain the correct solution. Thus...represented by simple inertial components with appropriate model connectivity instead to determine the free modal response of powertrain type

  13. Machine Learning and Deep Learning Models to Predict Runoff Water Quantity and Quality

    NASA Astrophysics Data System (ADS)

    Bradford, S. A.; Liang, J.; Li, W.; Murata, T.; Simunek, J.

    2017-12-01

    Contaminants can be rapidly transported at the soil surface by runoff to surface water bodies. Physically-based models, which are based on the mathematical description of main hydrological processes, are key tools for predicting surface water impairment. Along with physically-based models, data-driven models are becoming increasingly popular for describing the behavior of hydrological and water resources systems since these models can be used to complement or even replace physically based-models. In this presentation we propose a new data-driven model as an alternative to a physically-based overland flow and transport model. First, we have developed a physically-based numerical model to simulate overland flow and contaminant transport (the HYDRUS-1D overland flow module). A large number of numerical simulations were carried out to develop a database containing information about the impact of various input parameters (weather patterns, surface topography, vegetation, soil conditions, contaminants, and best management practices) on runoff water quantity and quality outputs. This database was used to train data-driven models. Three different methods (Neural Networks, Support Vector Machines, and Recurrence Neural Networks) were explored to prepare input- output functional relations. Results demonstrate the ability and limitations of machine learning and deep learning models to predict runoff water quantity and quality.

  14. Integration of Error Compensation of Coordinate Measuring Machines into Feature Measurement: Part I—Model Development

    PubMed Central

    Calvo, Roque; D’Amato, Roberto; Gómez, Emilio; Domingo, Rosario

    2016-01-01

    The development of an error compensation model for coordinate measuring machines (CMMs) and its integration into feature measurement is presented. CMMs are widespread and dependable instruments in industry and laboratories for dimensional measurement. From the tip probe sensor to the machine display, there is a complex transformation of probed point coordinates through the geometrical feature model that makes the assessment of accuracy and uncertainty measurement results difficult. Therefore, error compensation is not standardized, conversely to other simpler instruments. Detailed coordinate error compensation models are generally based on CMM as a rigid-body and it requires a detailed mapping of the CMM’s behavior. In this paper a new model type of error compensation is proposed. It evaluates the error from the vectorial composition of length error by axis and its integration into the geometrical measurement model. The non-explained variability by the model is incorporated into the uncertainty budget. Model parameters are analyzed and linked to the geometrical errors and uncertainty of CMM response. Next, the outstanding measurement models of flatness, angle, and roundness are developed. The proposed models are useful for measurement improvement with easy integration into CMM signal processing, in particular in industrial environments where built-in solutions are sought. A battery of implementation tests are presented in Part II, where the experimental endorsement of the model is included. PMID:27690052

  15. A Model-Free Machine Learning Method for Risk Classification and Survival Probability Prediction.

    PubMed

    Geng, Yuan; Lu, Wenbin; Zhang, Hao Helen

    2014-01-01

    Risk classification and survival probability prediction are two major goals in survival data analysis since they play an important role in patients' risk stratification, long-term diagnosis, and treatment selection. In this article, we propose a new model-free machine learning framework for risk classification and survival probability prediction based on weighted support vector machines. The new procedure does not require any specific parametric or semiparametric model assumption on data, and is therefore capable of capturing nonlinear covariate effects. We use numerous simulation examples to demonstrate finite sample performance of the proposed method under various settings. Applications to a glioma tumor data and a breast cancer gene expression survival data are shown to illustrate the new methodology in real data analysis.

  16. Abstract: Inference and Interval Estimation for Indirect Effects With Latent Variable Models.

    PubMed

    Falk, Carl F; Biesanz, Jeremy C

    2011-11-30

    Models specifying indirect effects (or mediation) and structural equation modeling are both popular in the social sciences. Yet relatively little research has compared methods that test for indirect effects among latent variables and provided precise estimates of the effectiveness of different methods. This simulation study provides an extensive comparison of methods for constructing confidence intervals and for making inferences about indirect effects with latent variables. We compared the percentile (PC) bootstrap, bias-corrected (BC) bootstrap, bias-corrected accelerated (BC a ) bootstrap, likelihood-based confidence intervals (Neale & Miller, 1997), partial posterior predictive (Biesanz, Falk, and Savalei, 2010), and joint significance tests based on Wald tests or likelihood ratio tests. All models included three reflective latent variables representing the independent, dependent, and mediating variables. The design included the following fully crossed conditions: (a) sample size: 100, 200, and 500; (b) number of indicators per latent variable: 3 versus 5; (c) reliability per set of indicators: .7 versus .9; (d) and 16 different path combinations for the indirect effect (α = 0, .14, .39, or .59; and β = 0, .14, .39, or .59). Simulations were performed using a WestGrid cluster of 1680 3.06GHz Intel Xeon processors running R and OpenMx. Results based on 1,000 replications per cell and 2,000 resamples per bootstrap method indicated that the BC and BC a bootstrap methods have inflated Type I error rates. Likelihood-based confidence intervals and the PC bootstrap emerged as methods that adequately control Type I error and have good coverage rates.

  17. Machine Learning

    SciTech Connect

    Chikkagoudar, Satish; Chatterjee, Samrat; Thomas, Dennis G.

    The absence of a robust and unified theory of cyber dynamics presents challenges and opportunities for using machine learning based data-driven approaches to further the understanding of the behavior of such complex systems. Analysts can also use machine learning approaches to gain operational insights. In order to be operationally beneficial, cybersecurity machine learning based models need to have the ability to: (1) represent a real-world system, (2) infer system properties, and (3) learn and adapt based on expert knowledge and observations. Probabilistic models and Probabilistic graphical models provide these necessary properties and are further explored in this chapter. Bayesian Networksmore » and Hidden Markov Models are introduced as an example of a widely used data driven classification/modeling strategy.« less

  18. A Multi-scale, Multi-Model, Machine-Learning Solar Forecasting Technology

    SciTech Connect

    Hamann, Hendrik F.

    The goal of the project was the development and demonstration of a significantly improved solar forecasting technology (short: Watt-sun), which leverages new big data processing technologies and machine-learnt blending between different models and forecast systems. The technology aimed demonstrating major advances in accuracy as measured by existing and new metrics which themselves were developed as part of this project. Finally, the team worked with Independent System Operators (ISOs) and utilities to integrate the forecasts into their operations.

  19. Modeling Physical Processes at the Nanoscale—Insight into Self-Organization of Small Systems (abstract)

    NASA Astrophysics Data System (ADS)

    Proykova, Ana

    2009-04-01

    Essential contributions have been made in the field of finite-size systems of ingredients interacting with potentials of various ranges. Theoretical simulations have revealed peculiar size effects on stability, ground state structure, phases, and phase transformation of systems confined in space and time. Models developed in the field of pure physics (atomic and molecular clusters) have been extended and successfully transferred to finite-size systems that seem very different—small-scale financial markets, autoimmune reactions, and social group reactions to advertisements. The models show that small-scale markets diverge unexpectedly fast as a result of small fluctuations; autoimmune reactions are sequences of two discontinuous phase transitions; and social groups possess critical behavior (social percolation) under the influence of an external field (advertisement). Some predicted size-dependent properties have been experimentally observed. These findings lead to the hypothesis that restrictions on an object's size determine the object's total internal (configuration) and external (environmental) interactions. Since phases are emergent phenomena produced by self-organization of a large number of particles, the occurrence of a phase in a system containing a small number of ingredients is remarkable.

  20. Study on intelligent processing system of man-machine interactive garment frame model

    NASA Astrophysics Data System (ADS)

    Chen, Shuwang; Yin, Xiaowei; Chang, Ruijiang; Pan, Peiyun; Wang, Xuedi; Shi, Shuze; Wei, Zhongqian

    2018-05-01

    A man-machine interactive garment frame model intelligent processing system is studied in this paper. The system consists of several sensor device, voice processing module, mechanical parts and data centralized acquisition devices. The sensor device is used to collect information on the environment changes brought by the body near the clothes frame model, the data collection device is used to collect the information of the environment change induced by the sensor device, voice processing module is used for speech recognition of nonspecific person to achieve human-machine interaction, mechanical moving parts are used to make corresponding mechanical responses to the information processed by data collection device.it is connected with data acquisition device by a means of one-way connection. There is a one-way connection between sensor device and data collection device, two-way connection between data acquisition device and voice processing module. The data collection device is one-way connection with mechanical movement parts. The intelligent processing system can judge whether it needs to interact with the customer, realize the man-machine interaction instead of the current rigid frame model.

  1. Mechanistic models versus machine learning, a fight worth fighting for the biological community?

    PubMed

    Baker, Ruth E; Peña, Jose-Maria; Jayamohan, Jayaratnam; Jérusalem, Antoine

    2018-05-01

    Ninety per cent of the world's data have been generated in the last 5 years ( Machine learning: the power and promise of computers that learn by example Report no. DES4702. Issued April 2017. Royal Society). A small fraction of these data is collected with the aim of validating specific hypotheses. These studies are led by the development of mechanistic models focused on the causality of input-output relationships. However, the vast majority is aimed at supporting statistical or correlation studies that bypass the need for causality and focus exclusively on prediction. Along these lines, there has been a vast increase in the use of machine learning models, in particular in the biomedical and clinical sciences, to try and keep pace with the rate of data generation. Recent successes now beg the question of whether mechanistic models are still relevant in this area. Said otherwise, why should we try to understand the mechanisms of disease progression when we can use machine learning tools to directly predict disease outcome? © 2018 The Author(s).

  2. Hierarchical analytical and simulation modelling of human-machine systems with interference

    NASA Astrophysics Data System (ADS)

    Braginsky, M. Ya; Tarakanov, D. V.; Tsapko, S. G.; Tsapko, I. V.; Baglaeva, E. A.

    2017-01-01

    The article considers the principles of building the analytical and simulation model of the human operator and the industrial control system hardware and software. E-networks as the extension of Petri nets are used as the mathematical apparatus. This approach allows simulating complex parallel distributed processes in human-machine systems. The structural and hierarchical approach is used as the building method for the mathematical model of the human operator. The upper level of the human operator is represented by the logical dynamic model of decision making based on E-networks. The lower level reflects psychophysiological characteristics of the human-operator.

  3. Abstraction and art.

    PubMed Central

    Gortais, Bernard

    2003-01-01

    In a given social context, artistic creation comprises a set of processes, which relate to the activity of the artist and the activity of the spectator. Through these processes we see and understand that the world is vaster than it is said to be. Artistic processes are mediated experiences that open up the world. A successful work of art expresses a reality beyond actual reality: it suggests an unknown world using the means and the signs of the known world. Artistic practices incorporate the means of creation developed by science and technology and change forms as they change. Artists and the public follow different processes of abstraction at different levels, in the definition of the means of creation, of representation and of perception of a work of art. This paper examines how the processes of abstraction are used within the framework of the visual arts and abstract painting, which appeared during a period of growing importance for the processes of abstraction in science and technology, at the beginning of the twentieth century. The development of digital platforms and new man-machine interfaces allow multimedia creations. This is performed under the constraint of phases of multidisciplinary conceptualization using generic representation languages, which tend to abolish traditional frontiers between the arts: visual arts, drama, dance and music. PMID:12903659

  4. Abstraction and art.

    PubMed

    Gortais, Bernard

    2003-07-29

    In a given social context, artistic creation comprises a set of processes, which relate to the activity of the artist and the activity of the spectator. Through these processes we see and understand that the world is vaster than it is said to be. Artistic processes are mediated experiences that open up the world. A successful work of art expresses a reality beyond actual reality: it suggests an unknown world using the means and the signs of the known world. Artistic practices incorporate the means of creation developed by science and technology and change forms as they change. Artists and the public follow different processes of abstraction at different levels, in the definition of the means of creation, of representation and of perception of a work of art. This paper examines how the processes of abstraction are used within the framework of the visual arts and abstract painting, which appeared during a period of growing importance for the processes of abstraction in science and technology, at the beginning of the twentieth century. The development of digital platforms and new man-machine interfaces allow multimedia creations. This is performed under the constraint of phases of multidisciplinary conceptualization using generic representation languages, which tend to abolish traditional frontiers between the arts: visual arts, drama, dance and music.

  5. Improving virtual screening predictive accuracy of Human kallikrein 5 inhibitors using machine learning models.

    PubMed

    Fang, Xingang; Bagui, Sikha; Bagui, Subhash

    2017-08-01

    The readily available high throughput screening (HTS) data from the PubChem database provides an opportunity for mining of small molecules in a variety of biological systems using machine learning techniques. From the thousands of available molecular descriptors developed to encode useful chemical information representing the characteristics of molecules, descriptor selection is an essential step in building an optimal quantitative structural-activity relationship (QSAR) model. For the development of a systematic descriptor selection strategy, we need the understanding of the relationship between: (i) the descriptor selection; (ii) the choice of the machine learning model; and (iii) the characteristics of the target bio-molecule. In this work, we employed the Signature descriptor to generate a dataset on the Human kallikrein 5 (hK 5) inhibition confirmatory assay data and compared multiple classification models including logistic regression, support vector machine, random forest and k-nearest neighbor. Under optimal conditions, the logistic regression model provided extremely high overall accuracy (98%) and precision (90%), with good sensitivity (65%) in the cross validation test. In testing the primary HTS screening data with more than 200K molecular structures, the logistic regression model exhibited the capability of eliminating more than 99.9% of the inactive structures. As part of our exploration of the descriptor-model-target relationship, the excellent predictive performance of the combination of the Signature descriptor and the logistic regression model on the assay data of the Human kallikrein 5 (hK 5) target suggested a feasible descriptor/model selection strategy on similar targets. Copyright © 2017 Elsevier Ltd. All rights reserved.

  6. Simple full micromagnetic model of exchange bias behavior in ferro/antiferromagnetic layered structures (abstract)

    NASA Astrophysics Data System (ADS)

    Koon, Norman C.

    1997-04-01

    It is shown using full micromagnetic relaxation calculations that exchange bias behavior is predicted for single-crystal ferro/antiferromagnetic layers with a fully compensated interface. The particular example most fully studied has a bcc/bct lattice structure with a fully compensated (110) interface plane. Only bilinear Heisenberg exchange was assumed, with anisotropy only in the antiferromagnet. In spite of the intuitive notion that exchange coupling between a ferromagnet and an antiferromagnet across a fully compensated plane of the antiferromagnet should be zero, we find strong coupling, comparable to the bilinear exchange, with a 90° angle between the ferromagnetic and antiferromagnetic axes of layers far from the interface in absence of an applied field. Even though the 90° coupling has characteristics resembling "biquadratic" exchange, it originates entirely from frustrated bilinear exchange. The development of exchange bias is found to originate from the formation of a domain wall in the antiferromagnet via the strong 90° exchange coupling and pinning of the wall by the magnetocrystalline anisotropy in the antiferromagnet. Because the large demagnetizing factor of the ferromagnet tends to confine its magnetization to the plane, the exchange bias is found to depend mainly on the strength and the symmetry of the in-plane component of anisotropy. Although little effort was made to analyze specific systems, the model reproduces many of the qualitative features observed in real exchange bias systems and gives reasonable semiquantitative estimates for the bias field when exchange and anisotropy values consistent with real systems are used.

  7. Learning with Technology: Video Modeling with Concrete-Representational-Abstract Sequencing for Students with Autism Spectrum Disorder.

    PubMed

    Yakubova, Gulnoza; Hughes, Elizabeth M; Shinaberry, Megan

    2016-07-01

    The purpose of this study was to determine the effectiveness of a video modeling intervention with concrete-representational-abstract instructional sequence in teaching mathematics concepts to students with autism spectrum disorder (ASD). A multiple baseline across skills design of single-case experimental methodology was used to determine the effectiveness of the intervention on the acquisition and maintenance of addition, subtraction, and number comparison skills for four elementary school students with ASD. Findings supported the effectiveness of the intervention in improving skill acquisition and maintenance at a 3-week follow-up. Implications for practice and future research are discussed.

  8. Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping.

    PubMed

    Shafizadeh-Moghadam, Hossein; Valavi, Roozbeh; Shahabi, Himan; Chapi, Kamran; Shirzadi, Ataollah

    2018-07-01

    In this research, eight individual machine learning and statistical models are implemented and compared, and based on their results, seven ensemble models for flood susceptibility assessment are introduced. The individual models included artificial neural networks, classification and regression trees, flexible discriminant analysis, generalized linear model, generalized additive model, boosted regression trees, multivariate adaptive regression splines, and maximum entropy, and the ensemble models were Ensemble Model committee averaging (EMca), Ensemble Model confidence interval Inferior (EMciInf), Ensemble Model confidence interval Superior (EMciSup), Ensemble Model to estimate the coefficient of variation (EMcv), Ensemble Model to estimate the mean (EMmean), Ensemble Model to estimate the median (EMmedian), and Ensemble Model based on weighted mean (EMwmean). The data set covered 201 flood events in the Haraz watershed (Mazandaran province in Iran) and 10,000 randomly selected non-occurrence points. Among the individual models, the Area Under the Receiver Operating Characteristic (AUROC), which showed the highest value, belonged to boosted regression trees (0.975) and the lowest value was recorded for generalized linear model (0.642). On the other hand, the proposed EMmedian resulted in the highest accuracy (0.976) among all models. In spite of the outstanding performance of some models, nevertheless, variability among the prediction of individual models was considerable. Therefore, to reduce uncertainty, creating more generalizable, more stable, and less sensitive models, ensemble forecasting approaches and in particular the EMmedian is recommended for flood susceptibility assessment. Copyright © 2018 Elsevier Ltd. All rights reserved.

  9. Developing a dengue forecast model using machine learning: A case study in China

    PubMed Central

    Zhang, Qin; Wang, Li; Xiao, Jianpeng; Zhang, Qingying; Luo, Ganfeng; Li, Zhihao; He, Jianfeng; Zhang, Yonghui; Ma, Wenjun

    2017-01-01

    Background In China, dengue remains an important public health issue with expanded areas and increased incidence recently. Accurate and timely forecasts of dengue incidence in China are still lacking. We aimed to use the state-of-the-art machine learning algorithms to develop an accurate predictive model of dengue. Methodology/Principal findings Weekly dengue cases, Baidu search queries and climate factors (mean temperature, relative humidity and rainfall) during 2011–2014 in Guangdong were gathered. A dengue search index was constructed for developing the predictive models in combination with climate factors. The observed year and week were also included in the models to control for the long-term trend and seasonality. Several machine learning algorithms, including the support vector regression (SVR) algorithm, step-down linear regression model, gradient boosted regression tree algorithm (GBM), negative binomial regression model (NBM), least absolute shrinkage and selection operator (LASSO) linear regression model and generalized additive model (GAM), were used as candidate models to predict dengue incidence. Performance and goodness of fit of the models were assessed using the root-mean-square error (RMSE) and R-squared measures. The residuals of the models were examined using the autocorrelation and partial autocorrelation function analyses to check the validity of the models. The models were further validated using dengue surveillance data from five other provinces. The epidemics during the last 12 weeks and the peak of the 2014 large outbreak were accurately forecasted by the SVR model selected by a cross-validation technique. Moreover, the SVR model had the consistently smallest prediction error rates for tracking the dynamics of dengue and forecasting the outbreaks in other areas in China. Conclusion and significance The proposed SVR model achieved a superior performance in comparison with other forecasting techniques assessed in this study. The findings

  10. Developing a dengue forecast model using machine learning: A case study in China.

    PubMed

    Guo, Pi; Liu, Tao; Zhang, Qin; Wang, Li; Xiao, Jianpeng; Zhang, Qingying; Luo, Ganfeng; Li, Zhihao; He, Jianfeng; Zhang, Yonghui; Ma, Wenjun

    2017-10-01

    In China, dengue remains an important public health issue with expanded areas and increased incidence recently. Accurate and timely forecasts of dengue incidence in China are still lacking. We aimed to use the state-of-the-art machine learning algorithms to develop an accurate predictive model of dengue. Weekly dengue cases, Baidu search queries and climate factors (mean temperature, relative humidity and rainfall) during 2011-2014 in Guangdong were gathered. A dengue search index was constructed for developing the predictive models in combination with climate factors. The observed year and week were also included in the models to control for the long-term trend and seasonality. Several machine learning algorithms, including the support vector regression (SVR) algorithm, step-down linear regression model, gradient boosted regression tree algorithm (GBM), negative binomial regression model (NBM), least absolute shrinkage and selection operator (LASSO) linear regression model and generalized additive model (GAM), were used as candidate models to predict dengue incidence. Performance and goodness of fit of the models were assessed using the root-mean-square error (RMSE) and R-squared measures. The residuals of the models were examined using the autocorrelation and partial autocorrelation function analyses to check the validity of the models. The models were further validated using dengue surveillance data from five other provinces. The epidemics during the last 12 weeks and the peak of the 2014 large outbreak were accurately forecasted by the SVR model selected by a cross-validation technique. Moreover, the SVR model had the consistently smallest prediction error rates for tracking the dynamics of dengue and forecasting the outbreaks in other areas in China. The proposed SVR model achieved a superior performance in comparison with other forecasting techniques assessed in this study. The findings can help the government and community respond early to dengue epidemics.

  11. A general electromagnetic excitation model for electrical machines considering the magnetic saturation and rub impact

    NASA Astrophysics Data System (ADS)

    Xu, Xueping; Han, Qinkai; Chu, Fulei

    2018-03-01

    The electromagnetic vibration of electrical machines with an eccentric rotor has been extensively investigated. However, magnetic saturation was often neglected. Moreover, the rub impact between the rotor and stator is inevitable when the amplitude of the rotor vibration exceeds the air-gap. This paper aims to propose a general electromagnetic excitation model for electrical machines. First, a general model which takes the magnetic saturation and rub impact into consideration is proposed and validated by the finite element method and reference. The dynamic equations of a Jeffcott rotor system with electromagnetic excitation and mass imbalance are presented. Then, the effects of pole-pair number and rubbing parameters on vibration amplitude are studied and approaches restraining the amplitude are put forward. Finally, the influences of mass eccentricity, resultant magnetomotive force (MMF), stiffness coefficient, damping coefficient, contact stiffness and friction coefficient on the stability of the rotor system are investigated through the Floquet theory, respectively. The amplitude jumping phenomenon is observed in a synchronous generator for different pole-pair numbers. The changes of design parameters can alter the stability states of the rotor system and the range of parameter values forms the zone of stability, which lays helpful suggestions for the design and application of the electrical machines.

  12. Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients

    NASA Astrophysics Data System (ADS)

    Tang, Jie; Liu, Rong; Zhang, Yue-Li; Liu, Mou-Ze; Hu, Yong-Fang; Shao, Ming-Jie; Zhu, Li-Jun; Xin, Hua-Wen; Feng, Gui-Wen; Shang, Wen-Jun; Meng, Xiang-Guang; Zhang, Li-Rong; Ming, Ying-Zi; Zhang, Wei

    2017-02-01

    Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the “derivation cohort” to develop dose-prediction algorithm, while the remaining 20% constituted the “validation cohort” to test the final selected algorithm. MLR, artificial neural network (ANN), regression tree (RT), multivariate adaptive regression splines (MARS), boosted regression tree (BRT), support vector regression (SVR), random forest regression (RFR), lasso regression (LAR) and Bayesian additive regression trees (BART) were applied and their performances were compared in this work. Among all the machine learning models, RT performed best in both derivation [0.71 (0.67-0.76)] and validation cohorts [0.73 (0.63-0.82)]. In addition, the ideal rate of RT was 4% higher than that of MLR. To our knowledge, this is the first study to use machine learning models to predict TSD, which will further facilitate personalized medicine in tacrolimus administration in the future.

  13. Modeling and Analysis of High Torque Density Transverse Flux Machines for Direct-Drive Applications

    NASA Astrophysics Data System (ADS)

    Hasan, Iftekhar

    Commercially available permanent magnet synchronous machines (PMSM) typically use rare-earth-based permanent magnets (PM). However, volatility and uncertainty associated with the supply and cost of rare-earth magnets have caused a push for increased research into the development of non-rare-earth based PM machines and reluctance machines. Compared to other PMSM topologies, the Transverse Flux Machine (TFM) is a promising candidate to get higher torque densities at low speed for direct-drive applications, using non-rare-earth based PMs. The TFMs can be designed with a very small pole pitch which allows them to attain higher force density than conventional radial flux machines (RFM) and axial flux machines (AFM). This dissertation presents the modeling, electromagnetic design, vibration analysis, and prototype development of a novel non-rare-earth based PM-TFM for a direct-drive wind turbine application. The proposed TFM addresses the issues of low power factor, cogging torque, and torque ripple during the electromagnetic design phase. An improved Magnetic Equivalent Circuit (MEC) based analytical model was developed as an alternative to the time-consuming 3D Finite Element Analysis (FEA) for faster electromagnetic analysis of the TFM. The accuracy and reliability of the MEC model were verified, both with 3D-FEA and experimental results. The improved MEC model was integrated with a Particle Swarm Optimization (PSO) algorithm to further enhance the capability of the analytical tool for performing rigorous optimization of performance-sensitive machine design parameters to extract the highest torque density for rated speed. A novel concept of integrating the rotary transformer within the proposed TFM design was explored to completely eliminate the use of magnets from the TFM. While keeping the same machine envelope, and without changing the stator or rotor cores, the primary and secondary of a rotary transformer were embedded into the double-sided TFM. The proposed

  14. Machine learning methods enable predictive modeling of antibody feature:function relationships in RV144 vaccinees.

    PubMed

    Choi, Ickwon; Chung, Amy W; Suscovich, Todd J; Rerks-Ngarm, Supachai; Pitisuttithum, Punnee; Nitayaphan, Sorachai; Kaewkungwal, Jaranit; O'Connell, Robert J; Francis, Donald; Robb, Merlin L; Michael, Nelson L; Kim, Jerome H; Alter, Galit; Ackerman, Margaret E; Bailey-Kellogg, Chris

    2015-04-01

    The adaptive immune response to vaccination or infection can lead to the production of specific antibodies to neutralize the pathogen or recruit innate immune effector cells for help. The non-neutralizing role of antibodies in stimulating effector cell responses may have been a key mechanism of the protection observed in the RV144 HIV vaccine trial. In an extensive investigation of a rich set of data collected from RV144 vaccine recipients, we here employ machine learning methods to identify and model associations between antibody features (IgG subclass and antigen specificity) and effector function activities (antibody dependent cellular phagocytosis, cellular cytotoxicity, and cytokine release). We demonstrate via cross-validation that classification and regression approaches can effectively use the antibody features to robustly predict qualitative and quantitative functional outcomes. This integration of antibody feature and function data within a machine learning framework provides a new, objective approach to discovering and assessing multivariate immune correlates.

  15. A model predictive current control of flux-switching permanent magnet machines for torque ripple minimization

    NASA Astrophysics Data System (ADS)

    Huang, Wentao; Hua, Wei; Yu, Feng

    2017-05-01

    Due to high airgap flux density generated by magnets and the special double salient structure, the cogging torque of the flux-switching permanent magnet (FSPM) machine is considerable, which limits the further applications. Based on the model predictive current control (MPCC) and the compensation control theory, a compensating-current MPCC (CC-MPCC) scheme is proposed and implemented to counteract the dominated components in cogging torque of an existing three-phase 12/10 FSPM prototyped machine, and thus to alleviate the influence of the cogging torque and improve the smoothness of electromagnetic torque as well as speed, where a comprehensive cost function is designed to evaluate the switching states. The simulated results indicate that the proposed CC-MPCC scheme can suppress the torque ripple significantly and offer satisfactory dynamic performances by comparisons with the conventional MPCC strategy. Finally, experimental results validate both the theoretical and simulated predictions.

  16. Machine Learning Methods Enable Predictive Modeling of Antibody Feature:Function Relationships in RV144 Vaccinees

    PubMed Central

    Choi, Ickwon; Chung, Amy W.; Suscovich, Todd J.; Rerks-Ngarm, Supachai; Pitisuttithum, Punnee; Nitayaphan, Sorachai; Kaewkungwal, Jaranit; O'Connell, Robert J.; Francis, Donald; Robb, Merlin L.; Michael, Nelson L.; Kim, Jerome H.; Alter, Galit; Ackerman, Margaret E.; Bailey-Kellogg, Chris

    2015-01-01

    The adaptive immune response to vaccination or infection can lead to the production of specific antibodies to neutralize the pathogen or recruit innate immune effector cells for help. The non-neutralizing role of antibodies in stimulating effector cell responses may have been a key mechanism of the protection observed in the RV144 HIV vaccine trial. In an extensive investigation of a rich set of data collected from RV144 vaccine recipients, we here employ machine learning methods to identify and model associations between antibody features (IgG subclass and antigen specificity) and effector function activities (antibody dependent cellular phagocytosis, cellular cytotoxicity, and cytokine release). We demonstrate via cross-validation that classification and regression approaches can effectively use the antibody features to robustly predict qualitative and quantitative functional outcomes. This integration of antibody feature and function data within a machine learning framework provides a new, objective approach to discovering and assessing multivariate immune correlates. PMID:25874406

  17. Kinetic modeling of α-hydrogen abstractions from unsaturated and saturated oxygenate compounds by carbon-centered radicals.

    PubMed

    Paraskevas, Paschalis D; Sabbe, Maarten K; Reyniers, Marie-Françoise; Papayannakos, Nikos; Marin, Guy B

    2014-06-23

    Hydrogen abstractions are important elementary reactions in a variety of reacting media at high temperatures in which oxygenates and hydrocarbon radicals are present. Accurate kinetic data are obtained from CBS-QB3 ab initio (AI) calculations by using conventional transition-state theory within the high-pressure limit, including corrections for hindered rotation and tunneling. From the obtained results, a group-additive (GA) model is developed that allows the Arrhenius parameters and rate coefficients for abstraction of the α-hydrogen from a wide range of oxygenate compounds to be predicted at temperatures ranging from 300 to 1500 K. From a training set of 60 hydrogen abstractions from oxygenates by carbon-centered radicals, 15 GA values (ΔGAV°s) are obtained for both the forward and reverse reactions. Among them, four ΔGAV°s refer to primary contributions, and the remaining 11 ΔGAV°s refer to secondary ones. The accuracy of the model is further improved by introducing seven corrections for cross-resonance stabilization of the transition state from an additional set of 43 reactions. The determined ΔGAV°s are validated upon a test set of AI data for 17 reactions. The mean absolute deviation of the pre-exponential factors (log A) and activation energies (E(a)) for the forward reaction at 300 K are 0.238 log(m(3)  mol(-1)  s(-1)) and 1.5 kJ mol(-1), respectively, whereas the mean factor of deviation <ρ> between the GA-predicted and the AI-calculated rate coefficients is 1.6. In comparison with a compilation of 33 experimental rate coefficients, the <ρ> between the GA-predicted values and these experimental values is only 2.2. Hence, the constructed GA model can be reliably used in the prediction of the kinetics of α-hydrogen-abstraction reactions between a broad range of oxygenates and oxygenate radicals. © 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

  18. ERGONOMICS ABSTRACTS 48347-48982.

    ERIC Educational Resources Information Center

    Ministry of Technology, London (England). Warren Spring Lab.

    IN THIS COLLECTION OF ERGONOMICS ABSTRACTS AND ANNOTATIONS THE FOLLOWING AREAS OF CONCERN ARE REPRESENTED--GENERAL REFERENCES, METHODS, FACILITIES, AND EQUIPMENT RELATING TO ERGONOMICS, SYSTEMS OF MAN AND MACHINES, VISUAL, AUDITORY, AND OTHER SENSORY INPUTS AND PROCESSES (INCLUDING SPEECH AND INTELLIGIBILITY), INPUT CHANNELS, BODY MEASUREMENTS,…

  19. Predicting Pre-planting Risk of Stagonospora nodorum blotch in Winter Wheat Using Machine Learning Models.

    PubMed

    Mehra, Lucky K; Cowger, Christina; Gross, Kevin; Ojiambo, Peter S

    2016-01-01

    Pre-planting factors have been associated with the late-season severity of Stagonospora nodorum blotch (SNB), caused by the fungal pathogen Parastagonospora nodorum, in winter wheat (Triticum aestivum). The relative importance of these factors in the risk of SNB has not been determined and this knowledge can facilitate disease management decisions prior to planting of the wheat crop. In this study, we examined the performance of multiple regression (MR) and three machine learning algorithms namely artificial neural networks, categorical and regression trees, and random forests (RF), in predicting the pre-planting risk of SNB in wheat. Pre-planting factors tested as potential predictor variables were cultivar resistance, latitude, longitude, previous crop, seeding rate, seed treatment, tillage type, and wheat residue. Disease severity assessed at the end of the growing season was used as the response variable. The models were developed using 431 disease cases (unique combinations of predictors) collected from 2012 to 2014 and these cases were randomly divided into training, validation, and test datasets. Models were evaluated based on the regression of observed against predicted severity values of SNB, sensitivity-specificity ROC analysis, and the Kappa statistic. A strong relationship was observed between late-season severity of SNB and specific pre-planting factors in which latitude, longitude, wheat residue, and cultivar resistance were the most important predictors. The MR model explained 33% of variability in the data, while machine learning models explained 47 to 79% of the total variability. Similarly, the MR model correctly classified 74% of the disease cases, while machine learning models correctly classified 81 to 83% of these cases. Results show that the RF algorithm, which explained 79% of the variability within the data, was the most accurate in predicting the risk of SNB, with an accuracy rate of 93%. The RF algorithm could allow early assessment of

  20. Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems

    DOE PAGES

    Trehan, Sumeet; Carlberg, Kevin T.; Durlofsky, Louis J.

    2017-07-14

    A machine learning–based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (eg, random forests, and LASSO) to map a large set of inexpensively computed “error indicators” (ie, features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed bymore » simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering) and subsequently constructs a “local” regression model to predict the time-instantaneous error within each identified region of feature space. We consider 2 uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (eg, time-integrated errors). We then apply the proposed framework to model errors in reduced-order models of nonlinear oil-water subsurface flow simulations, with time-varying well-control (bottom-hole pressure) parameters. The reduced-order models used in this work entail application of trajectory piecewise linearization in conjunction with proper orthogonal decomposition. Moreover, when the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and

  1. Error modeling for surrogates of dynamical systems using machine learning: Machine-learning-based error model for surrogates of dynamical systems

    SciTech Connect

    Trehan, Sumeet; Carlberg, Kevin T.; Durlofsky, Louis J.

    A machine learning–based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (eg, random forests, and LASSO) to map a large set of inexpensively computed “error indicators” (ie, features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed bymore » simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering) and subsequently constructs a “local” regression model to predict the time-instantaneous error within each identified region of feature space. We consider 2 uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (eg, time-integrated errors). We then apply the proposed framework to model errors in reduced-order models of nonlinear oil-water subsurface flow simulations, with time-varying well-control (bottom-hole pressure) parameters. The reduced-order models used in this work entail application of trajectory piecewise linearization in conjunction with proper orthogonal decomposition. Moreover, when the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and

  2. Abstraction Techniques for Parameterized Verification

    DTIC Science & Technology

    2006-11-01

    approach for applying model checking to unbounded systems is to extract finite state models from them using conservative abstraction techniques. Prop...36 2.5.1 Multiple Reference Processes . . . . . . . . . . . . . . . . . . . 36 2.5.2 Adding Monitor Processes...model checking to complex pieces of code like device drivers depends on the use of abstraction methods. An abstraction method extracts a small finite

  3. Mathematical concepts for modeling human behavior in complex man-machine systems

    NASA Technical Reports Server (NTRS)

    Johannsen, G.; Rouse, W. B.

    1979-01-01

    Many human behavior (e.g., manual control) models have been found to be inadequate for describing processes in certain real complex man-machine systems. An attempt is made to find a way to overcome this problem by examining the range of applicability of existing mathematical models with respect to the hierarchy of human activities in real complex tasks. Automobile driving is chosen as a baseline scenario, and a hierarchy of human activities is derived by analyzing this task in general terms. A structural description leads to a block diagram and a time-sharing computer analogy.

  4. Constructing and validating readability models: the method of integrating multilevel linguistic features with machine learning.

    PubMed

    Sung, Yao-Ting; Chen, Ju-Ling; Cha, Ji-Her; Tseng, Hou-Chiang; Chang, Tao-Hsing; Chang, Kuo-En

    2015-06-01

    Multilevel linguistic features have been proposed for discourse analysis, but there have been few applications of multilevel linguistic features to readability models and also few validations of such models. Most traditional readability formulae are based on generalized linear models (GLMs; e.g., discriminant analysis and multiple regression), but these models have to comply with certain statistical assumptions about data properties and include all of the data in formulae construction without pruning the outliers in advance. The use of such readability formulae tends to produce a low text classification accuracy, while using a support vector machine (SVM) in machine learning can enhance the classification outcome. The present study constructed readability models by integrating multilevel linguistic features with SVM, which is more appropriate for text classification. Taking the Chinese language as an example, this study developed 31 linguistic features as the predicting variables at the word, semantic, syntax, and cohesion levels, with grade levels of texts as the criterion variable. The study compared four types of readability models by integrating unilevel and multilevel linguistic features with GLMs and an SVM. The results indicate that adopting a multilevel approach in readability analysis provides a better representation of the complexities of both texts and the reading comprehension process.

  5. Assessment of multi-wildfire occurrence data for machine learning based risk modelling

    NASA Astrophysics Data System (ADS)

    Lim, C. H.; Kim, M.; Kim, S. J.; Yoo, S.; Lee, W. K.

    2017-12-01

    The occurrence of East Asian wildfires is mainly caused by human-activities, but the extreme drought increased due to the climate change caused wildfires and they spread to large-scale fires. Accurate occurrence location data is required for modelling wildfire probability and risk. In South Korea, occurrence data surveyed through KFS (Korea Forest Service) and MODIS (MODerate-resolution Imaging Spectroradiometer) satellite-based active fire data can be utilized. In this study, two sorts of wildfire occurrence data were applied to select suitable occurrence data for machine learning based wildfire risk modelling. MaxEnt (Maximum Entropy) model based on machine learning is used for wildfire risk modelling, and two types of occurrence data and socio-economic and climate-environment data are applied to modelling. In the results with KFS survey based data, the low relationship was shown with climate-environmental factors, and the uncertainty of coordinate information appeared. The MODIS-based active fire data were found outside the forests, and there were a lot of spots that did not match the actual wildfires. In order to utilize MODIS-based active fire data, it was necessary to extract forest area and utilize only high-confidence level data. In KFS data, it was necessary to separate the analysis according to the damage scale to improve the modelling accuracy. Ultimately, it is considered to be the best way to simulate the wildfire risk by constructing more accurate information by combining two sorts of wildfire occurrence data.

  6. A hybrid prognostic model for multistep ahead prediction of machine condition

    NASA Astrophysics Data System (ADS)

    Roulias, D.; Loutas, T. H.; Kostopoulos, V.

    2012-05-01

    Prognostics are the future trend in condition based maintenance. In the current framework a data driven prognostic model is developed. The typical procedure of developing such a model comprises a) the selection of features which correlate well with the gradual degradation of the machine and b) the training of a mathematical tool. In this work the data are taken from a laboratory scale single stage gearbox under multi-sensor monitoring. Tests monitoring the condition of the gear pair from healthy state until total brake down following several days of continuous operation were conducted. After basic pre-processing of the derived data, an indicator that correlated well with the gearbox condition was obtained. Consecutively the time series is split in few distinguishable time regions via an intelligent data clustering scheme. Each operating region is modelled with a feed-forward artificial neural network (FFANN) scheme. The performance of the proposed model is tested by applying the system to predict the machine degradation level on unseen data. The results show the plausibility and effectiveness of the model in following the trend of the timeseries even in the case that a sudden change occurs. Moreover the model shows ability to generalise for application in similar mechanical assets.

  7. Exploring the influence of constitutive models and associated parameters for the orthogonal machining of Ti6Al4V

    NASA Astrophysics Data System (ADS)

    Pervaiz, S.; Anwar, S.; Kannan, S.; Almarfadi, A.

    2018-04-01

    Ti6Al4V is known as difficult-to-cut material due to its inherent properties such as high hot hardness, low thermal conductivity and high chemical reactivity. Though, Ti6Al4V is utilized by industrial sectors such as aeronautics, energy generation, petrochemical and bio-medical etc. For the metal cutting community, competent and cost-effective machining of Ti6Al4V is a challenging task. To optimize cost and machining performance for the machining of Ti6Al4V, finite element based cutting simulation can be a very useful tool. The aim of this paper is to develop a finite element machining model for the simulation of Ti6Al4V machining process. The study incorporates material constitutive models namely Power Law (PL) and Johnson – Cook (JC) material models to mimic the mechanical behaviour of Ti6Al4V. The study investigates cutting temperatures, cutting forces, stresses, and plastic strains with respect to different PL and JC material models with associated parameters. In addition, the numerical study also integrates different cutting tool rake angles in the machining simulations. The simulated results will be beneficial to draw conclusions for improving the overall machining performance of Ti6Al4V.

  8. Support Vector Machines Model of Computed Tomography for Assessing Lymph Node Metastasis in Esophageal Cancer with Neoadjuvant Chemotherapy.

    PubMed

    Wang, Zhi-Long; Zhou, Zhi-Guo; Chen, Ying; Li, Xiao-Ting; Sun, Ying-Shi

    The aim of this study was to diagnose lymph node metastasis of esophageal cancer by support vector machines model based on computed tomography. A total of 131 esophageal cancer patients with preoperative chemotherapy and radical surgery were included. Various indicators (tumor thickness, tumor length, tumor CT value, total number of lymph nodes, and long axis and short axis sizes of largest lymph node) on CT images before and after neoadjuvant chemotherapy were recorded. A support vector machines model based on these CT indicators was built to predict lymph node metastasis. Support vector machines model diagnosed lymph node metastasis better than preoperative short axis size of largest lymph node on CT. The area under the receiver operating characteristic curves were 0.887 and 0.705, respectively. The support vector machine model of CT images can help diagnose lymph node metastasis in esophageal cancer with preoperative chemotherapy.

  9. A Critical Review for Developing Accurate and Dynamic Predictive Models Using Machine Learning Methods in Medicine and Health Care.

    PubMed

    Alanazi, Hamdan O; Abdullah, Abdul Hanan; Qureshi, Kashif Naseer

    2017-04-01

    Recently, Artificial Intelligence (AI) has been used widely in medicine and health care sector. In machine learning, the classification or prediction is a major field of AI. Today, the study of existing predictive models based on machine learning methods is extremely active. Doctors need accurate predictions for the outcomes of their patients' diseases. In addition, for accurate predictions, timing is another significant factor that influences treatment decisions. In this paper, existing predictive models in medicine and health care have critically reviewed. Furthermore, the most famous machine learning methods have explained, and the confusion between a statistical approach and machine learning has clarified. A review of related literature reveals that the predictions of existing predictive models differ even when the same dataset is used. Therefore, existing predictive models are essential, and current methods must be improved.

  10. Operating Comfort Prediction Model of Human-Machine Interface Layout for Cabin Based on GEP.

    PubMed

    Deng, Li; Wang, Guohua; Chen, Bo

    2015-01-01

    In view of the evaluation and decision-making problem of human-machine interface layout design for cabin, the operating comfort prediction model is proposed based on GEP (Gene Expression Programming), using operating comfort to evaluate layout scheme. Through joint angles to describe operating posture of upper limb, the joint angles are taken as independent variables to establish the comfort model of operating posture. Factor analysis is adopted to decrease the variable dimension; the model's input variables are reduced from 16 joint angles to 4 comfort impact factors, and the output variable is operating comfort score. The Chinese virtual human body model is built by CATIA software, which will be used to simulate and evaluate the operators' operating comfort. With 22 groups of evaluation data as training sample and validation sample, GEP algorithm is used to obtain the best fitting function between the joint angles and the operating comfort; then, operating comfort can be predicted quantitatively. The operating comfort prediction result of human-machine interface layout of driller control room shows that operating comfort prediction model based on GEP is fast and efficient, it has good prediction effect, and it can improve the design efficiency.

  11. A comparison of optimal MIMO linear and nonlinear models for brain machine interfaces

    NASA Astrophysics Data System (ADS)

    Kim, S.-P.; Sanchez, J. C.; Rao, Y. N.; Erdogmus, D.; Carmena, J. M.; Lebedev, M. A.; Nicolelis, M. A. L.; Principe, J. C.

    2006-06-01

    The field of brain-machine interfaces requires the estimation of a mapping from spike trains collected in motor cortex areas to the hand kinematics of the behaving animal. This paper presents a systematic investigation of several linear (Wiener filter, LMS adaptive filters, gamma filter, subspace Wiener filters) and nonlinear models (time-delay neural network and local linear switching models) applied to datasets from two experiments in monkeys performing motor tasks (reaching for food and target hitting). Ensembles of 100-200 cortical neurons were simultaneously recorded in these experiments, and even larger neuronal samples are anticipated in the future. Due to the large size of the models (thousands of parameters), the major issue studied was the generalization performance. Every parameter of the models (not only the weights) was selected optimally using signal processing and machine learning techniques. The models were also compared statistically with respect to the Wiener filter as the baseline. Each of the optimization procedures produced improvements over that baseline for either one of the two datasets or both.

  12. Operating Comfort Prediction Model of Human-Machine Interface Layout for Cabin Based on GEP

    PubMed Central

    Wang, Guohua; Chen, Bo

    2015-01-01

    In view of the evaluation and decision-making problem of human-machine interface layout design for cabin, the operating comfort prediction model is proposed based on GEP (Gene Expression Programming), using operating comfort to evaluate layout scheme. Through joint angles to describe operating posture of upper limb, the joint angles are taken as independent variables to establish the comfort model of operating posture. Factor analysis is adopted to decrease the variable dimension; the model's input variables are reduced from 16 joint angles to 4 comfort impact factors, and the output variable is operating comfort score. The Chinese virtual human body model is built by CATIA software, which will be used to simulate and evaluate the operators' operating comfort. With 22 groups of evaluation data as training sample and validation sample, GEP algorithm is used to obtain the best fitting function between the joint angles and the operating comfort; then, operating comfort can be predicted quantitatively. The operating comfort prediction result of human-machine interface layout of driller control room shows that operating comfort prediction model based on GEP is fast and efficient, it has good prediction effect, and it can improve the design efficiency. PMID:26448740

  13. Estimating Inflows to Lake Okeechobee Using Climate Indices: A Machine Learning Modeling Approach

    NASA Astrophysics Data System (ADS)

    Kalra, A.; Ahmad, S.

    2008-12-01

    The operation of regional water management systems that include lakes and storage reservoirs for flood control and water supply can be significantly improved by using climate indices. This research is focused on forecasting Lag 1 annual inflow to Lake Okeechobee, located in South Florida, using annual oceanic- atmospheric indices of Pacific Decadal Oscillation (PDO), North Atlantic Oscillation (NAO), Atlantic Multidecadal Oscillation (AMO), and El Nino-Southern Oscillations (ENSO). Support Vector Machine (SVM) and Least Square Support Vector Machine (LSSVM), belonging to the class of data driven models, are developed to forecast annual lake inflow using annual oceanic-atmospheric indices data from 1914 to 2003. The models were trained with 80 years of data and tested for 10 years of data. Based on Correlation Coefficient, Root Means Square Error, and Mean Absolute Error model predictions were in good agreement with measured inflow volumes. Sensitivity analysis, performed to evaluate the effect of individual and coupled oscillations, revealed a strong signal for AMO and ENSO indices compared to PDO and NAO indices for one year lead-time inflow forecast. Inflow predictions from the SVM models were better when compared with the predictions obtained from feed forward back propagation Artificial Neural Network (ANN) models.

  14. A comparison of optimal MIMO linear and nonlinear models for brain-machine interfaces.

    PubMed

    Kim, S-P; Sanchez, J C; Rao, Y N; Erdogmus, D; Carmena, J M; Lebedev, M A; Nicolelis, M A L; Principe, J C

    2006-06-01

    The field of brain-machine interfaces requires the estimation of a mapping from spike trains collected in motor cortex areas to the hand kinematics of the behaving animal. This paper presents a systematic investigation of several linear (Wiener filter, LMS adaptive filters, gamma filter, subspace Wiener filters) and nonlinear models (time-delay neural network and local linear switching models) applied to datasets from two experiments in monkeys performing motor tasks (reaching for food and target hitting). Ensembles of 100-200 cortical neurons were simultaneously recorded in these experiments, and even larger neuronal samples are anticipated in the future. Due to the large size of the models (thousands of parameters), the major issue studied was the generalization performance. Every parameter of the models (not only the weights) was selected optimally using signal processing and machine learning techniques. The models were also compared statistically with respect to the Wiener filter as the baseline. Each of the optimization procedures produced improvements over that baseline for either one of the two datasets or both.

  15. [Research, design and application of model NSE-1 neck muscle training machine for pilots].

    PubMed

    Cheng, Haiping; Wang, Zhijie; Liu, Songyang; Yang, Yi; Zhao, Guang; Cong, Hong; Han, Xueping; Liu, Min; Yu, Mengsun

    2011-04-01

    Pain in the cervical region of air force pilots, who are exposed to high G-forces, is a specifically occupational health problem. To minimize neck problems, the cervical muscles need specific strength exercise. It is important that the training for the neck must be carried out with optimal resistance in exercises. The model NSE-1 neck training machine for pilots was designed for neck strengthening exercises under safe and effective conditions. In order to realize the functions of changeable velocity and resistant (CVR) training and neck isometric contractive exercises, the techniques of adaptive hydraulics, sensor, optic and auditory biological feedback, and signal processing were applied to this machine. The training system mainly consists of mechanical parts (including the chair of flexion and extension, the chair of right and left lateral flexion, the components of hydraulics and torque transformer, etc.), and the software of signal processing and biological feedback. Eleven volunteers were selected for the experiments of neck isometric contractive exercises, three times a week for 6 weeks, where CVR training (flexion, extension, right, left lateral flexion) one time a week. The increase in relative strength of the neck (flexion, extension, left and right lateral flexion) was 70.8%, 83.7%, 78.6% and 75.2%, respectively after training. Results show that the strength of the neck can be increased safely, effectively and rapidly with NSE-1 neck training machine to perform neck training.

  16. Solar Flare Prediction Model with Three Machine-learning Algorithms using Ultraviolet Brightening and Vector Magnetograms

    NASA Astrophysics Data System (ADS)

    Nishizuka, N.; Sugiura, K.; Kubo, Y.; Den, M.; Watari, S.; Ishii, M.

    2017-02-01

    We developed a flare prediction model using machine learning, which is optimized to predict the maximum class of flares occurring in the following 24 hr. Machine learning is used to devise algorithms that can learn from and make decisions on a huge amount of data. We used solar observation data during the period 2010-2015, such as vector magnetograms, ultraviolet (UV) emission, and soft X-ray emission taken by the Solar Dynamics Observatory and the Geostationary Operational Environmental Satellite. We detected active regions (ARs) from the full-disk magnetogram, from which ˜60 features were extracted with their time differentials, including magnetic neutral lines, the current helicity, the UV brightening, and the flare history. After standardizing the feature database, we fully shuffled and randomly separated it into two for training and testing. To investigate which algorithm is best for flare prediction, we compared three machine-learning algorithms: the support vector machine, k-nearest neighbors (k-NN), and extremely randomized trees. The prediction score, the true skill statistic, was higher than 0.9 with a fully shuffled data set, which is higher than that for human forecasts. It was found that k-NN has the highest performance among the three algorithms. The ranking of the feature importance showed that previous flare activity is most effective, followed by the length of magnetic neutral lines, the unsigned magnetic flux, the area of UV brightening, and the time differentials of features over 24 hr, all of which are strongly correlated with the flux emergence dynamics in an AR.

  17. Solar Flare Prediction Model with Three Machine-learning Algorithms using Ultraviolet Brightening and Vector Magnetograms

    SciTech Connect

    Nishizuka, N.; Kubo, Y.; Den, M.

    We developed a flare prediction model using machine learning, which is optimized to predict the maximum class of flares occurring in the following 24 hr. Machine learning is used to devise algorithms that can learn from and make decisions on a huge amount of data. We used solar observation data during the period 2010–2015, such as vector magnetograms, ultraviolet (UV) emission, and soft X-ray emission taken by the Solar Dynamics Observatory and the Geostationary Operational Environmental Satellite . We detected active regions (ARs) from the full-disk magnetogram, from which ∼60 features were extracted with their time differentials, including magnetic neutralmore » lines, the current helicity, the UV brightening, and the flare history. After standardizing the feature database, we fully shuffled and randomly separated it into two for training and testing. To investigate which algorithm is best for flare prediction, we compared three machine-learning algorithms: the support vector machine, k-nearest neighbors (k-NN), and extremely randomized trees. The prediction score, the true skill statistic, was higher than 0.9 with a fully shuffled data set, which is higher than that for human forecasts. It was found that k-NN has the highest performance among the three algorithms. The ranking of the feature importance showed that previous flare activity is most effective, followed by the length of magnetic neutral lines, the unsigned magnetic flux, the area of UV brightening, and the time differentials of features over 24 hr, all of which are strongly correlated with the flux emergence dynamics in an AR.« less

  18. Machine learning of atmospheric chemistry. Applications to a global chemistry transport model.

    NASA Astrophysics Data System (ADS)

    Evans, M. J.; Keller, C. A.

    2017-12-01

    Atmospheric chemistry is central to many environmental issues such as air pollution, climate change, and stratospheric ozone loss. Chemistry Transport Models (CTM) are a central tool for understanding these issues, whether for research or for forecasting. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (mili-seconds for O(1D) to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a CTM.We have investigated a machine learning approach to solving the differential equations instead of solving them numerically. From an annual simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (random regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry.This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, centennial scale climate simulations etc. We discuss our approches' speed and accuracy, and highlight some potential future directions for

  19. Feature combination networks for the interpretation of statistical machine learning models: application to Ames mutagenicity.

    PubMed

    Webb, Samuel J; Hanser, Thierry; Howlin, Brendan; Krause, Paul; Vessey, Jonathan D

    2014-03-25

    A new algorithm has been developed to enable the interpretation of black box models. The developed algorithm is agnostic to learning algorithm and open to all structural based descriptors such as fragments, keys and hashed fingerprints. The algorithm has provided meaningful interpretation of Ames mutagenicity predictions from both random forest and support vector machine models built on a variety of structural fingerprints.A fragmentation algorithm is utilised to investigate the model's behaviour on specific substructures present in the query. An output is formulated summarising causes of activation and deactivation. The algorithm is able to identify multiple causes of activation or deactivation in addition to identifying localised deactivations where the prediction for the query is active overall. No loss in performance is seen as there is no change in the prediction; the interpretation is produced directly on the model's behaviour for the specific query. Models have been built using multiple learning algorithms including support vector machine and random forest. The models were built on public Ames mutagenicity data and a variety of fingerprint descriptors were used. These models produced a good performance in both internal and external validation with accuracies around 82%. The models were used to evaluate the interpretation algorithm. Interpretation was revealed that links closely with understood mechanisms for Ames mutagenicity. This methodology allows for a greater utilisation of the predictions made by black box models and can expedite further study based on the output for a (quantitative) structure activity model. Additionally the algorithm could be utilised for chemical dataset investigation and knowledge extraction/human SAR development.

  20. Modeling and Designing of A Nonlineartemperature-Humidity Controller Using Inmushroom-Drying Machine

    NASA Astrophysics Data System (ADS)

    Wu, Xiuhua; Luo, Haiyan; Shi, Minhui

    Drying-process of many kinds of farm produce in a close room, such as mushroom-drying machine, is generally a complicated nonlinear and timedelay cause, in which the temperature and the humidity are the main controlled elements. The accurate controlling of the temperature and humidity is always an interesting problem. It's difficult and very important to make a more accurate mathematical model about the varying of the two. A math model was put forward after considering many aspects and analyzing the actual working circumstance in this paper. Form the model it can be seen that the changes of temperature and humidity in drying machine are not simple linear but an affine nonlinear process. Controlling the process exactly is the key that influences the quality of the dried mushroom. In this paper, the differential geometry theories and methods are used to analyze and solve the model of these smallenvironment elements. And at last a kind of nonlinear controller which satisfied the optimal quadratic performance index is designed. It can be proved more feasible and practical than the conventional controlling.

  1. Machine learning of frustrated classical spin models. I. Principal component analysis

    NASA Astrophysics Data System (ADS)

    Wang, Ce; Zhai, Hui

    2017-10-01

    This work aims at determining whether artificial intelligence can recognize a phase transition without prior human knowledge. If this were successful, it could be applied to, for instance, analyzing data from the quantum simulation of unsolved physical models. Toward this goal, we first need to apply the machine learning algorithm to well-understood models and see whether the outputs are consistent with our prior knowledge, which serves as the benchmark for this approach. In this work, we feed the computer data generated by the classical Monte Carlo simulation for the X Y model in frustrated triangular and union jack lattices, which has two order parameters and exhibits two phase transitions. We show that the outputs of the principal component analysis agree very well with our understanding of different orders in different phases, and the temperature dependences of the major components detect the nature and the locations of the phase transitions. Our work offers promise for using machine learning techniques to study sophisticated statistical models, and our results can be further improved by using principal component analysis with kernel tricks and the neural network method.

  2. Mathematical model of simple spalling formation during coal cutting with extracting machine

    NASA Astrophysics Data System (ADS)

    Gabov, V. V.; Zadkov, D. A.

    2018-05-01

    A single-mass model of a rotor shearer is analyzed. It is shown that rotor mining machines has large inertia moments and load dynamics. An extraction module model with selective movement of the cutting tool is represented. The peculiar feature of such extracting machines is fluid power drive cutter mechanism. They can steadily operate at large shear thickness, and locking modes are not an emergency for them. Comparing with shearers they have less inertional mass, but slower average cutting speed, and its momentary values depend on load. Basing on the equation of hydraulic fuel consumption balance the work of fluid power drive of extracting module cutter mechanism together with hydro pneumatic accumulator is analyzed. Spalling formation model during coal cutting with fluid power drive cutter mechanism and potential energy stores are suggested. Matching cutter speed with the speed of main crack expansion and amount of potential energy consumption, cutter load is determined only by ultimate stress at crack pole and friction. Tests of an extracting module cutter in real size model proved the stated theory.

  3. A Genetic Algorithm Based Support Vector Machine Model for Blood-Brain Barrier Penetration Prediction

    PubMed Central

    Zhang, Daqing; Xiao, Jianfeng; Zhou, Nannan; Luo, Xiaomin; Jiang, Hualiang; Chen, Kaixian

    2015-01-01

    Blood-brain barrier (BBB) is a highly complex physical barrier determining what substances are allowed to enter the brain. Support vector machine (SVM) is a kernel-based machine learning method that is widely used in QSAR study. For a successful SVM model, the kernel parameters for SVM and feature subset selection are the most important factors affecting prediction accuracy. In most studies, they are treated as two independent problems, but it has been proven that they could affect each other. We designed and implemented genetic algorithm (GA) to optimize kernel parameters and feature subset selection for SVM regression and applied it to the BBB penetration prediction. The results show that our GA/SVM model is more accurate than other currently available log BB models. Therefore, to optimize both SVM parameters and feature subset simultaneously with genetic algorithm is a better approach than other methods that treat the two problems separately. Analysis of our log BB model suggests that carboxylic acid group, polar surface area (PSA)/hydrogen-bonding ability, lipophilicity, and molecular charge play important role in BBB penetration. Among those properties relevant to BBB penetration, lipophilicity could enhance the BBB penetration while all the others are negatively correlated with BBB penetration. PMID:26504797

  4. Improved equivalent magnetic network modeling for analyzing working points of PMs in interior permanent magnet machine

    NASA Astrophysics Data System (ADS)

    Guo, Liyan; Xia, Changliang; Wang, Huimin; Wang, Zhiqiang; Shi, Tingna

    2018-05-01

    As is well known, the armature current will be ahead of the back electromotive force (back-EMF) under load condition of the interior permanent magnet (PM) machine. This kind of advanced armature current will produce a demagnetizing field, which may make irreversible demagnetization appeared in PMs easily. To estimate the working points of PMs more accurately and take demagnetization under consideration in the early design stage of a machine, an improved equivalent magnetic network model is established in this paper. Each PM under each magnetic pole is segmented, and the networks in the rotor pole shoe are refined, which makes a more precise model of the flux path in the rotor pole shoe possible. The working point of each PM under each magnetic pole can be calculated accurately by the established improved equivalent magnetic network model. Meanwhile, the calculated results are compared with those calculated by FEM. And the effects of d-axis component and q-axis component of armature current, air-gap length and flux barrier size on working points of PMs are analyzed by the improved equivalent magnetic network model.

  5. The applications of machine learning algorithms in the modeling of estrogen-like chemicals.

    PubMed

    Liu, Huanxiang; Yao, Xiaojun; Gramatica, Paola

    2009-06-01

    Increasing concern is being shown by the scientific community, government regulators, and the public about endocrine-disrupting chemicals that, in the environment, are adversely affecting human and wildlife health through a variety of mechanisms, mainly estrogen receptor-mediated mechanisms of toxicity. Because of the large number of such chemicals in the environment, there is a great need for an effective means of rapidly assessing endocrine-disrupting activity in the toxicology assessment process. When faced with the challenging task of screening large libraries of molecules for biological activity, the benefits of computational predictive models based on quantitative structure-activity relationships to identify possible estrogens become immediately obvious. Recently, in order to improve the accuracy of prediction, some machine learning techniques were introduced to build more effective predictive models. In this review we will focus our attention on some recent advances in the use of these methods in modeling estrogen-like chemicals. The advantages and disadvantages of the machine learning algorithms used in solving this problem, the importance of the validation and performance assessment of the built models as well as their applicability domains will be discussed.

  6. Integrating Machine Learning into a Crowdsourced Model for Earthquake-Induced Damage Assessment

    NASA Technical Reports Server (NTRS)

    Rebbapragada, Umaa; Oommen, Thomas

    2011-01-01

    On January 12th, 2010, a catastrophic 7.0M earthquake devastated the country of Haiti. In the aftermath of an earthquake, it is important to rapidly assess damaged areas in order to mobilize the appropriate resources. The Haiti damage assessment effort introduced a promising model that uses crowdsourcing to map damaged areas in freely available remotely-sensed data. This paper proposes the application of machine learning methods to improve this model. Specifically, we apply work on learning from multiple, imperfect experts to the assessment of volunteer reliability, and propose the use of image segmentation to automate the detection of damaged areas. We wrap both tasks in an active learning framework in order to shift volunteer effort from mapping a full catalog of images to the generation of high-quality training data. We hypothesize that the integration of machine learning into this model improves its reliability, maintains the speed of damage assessment, and allows the model to scale to higher data volumes.

  7. A model-based analysis of impulsivity using a slot-machine gambling paradigm

    PubMed Central

    Paliwal, Saee; Petzschner, Frederike H.; Schmitz, Anna Katharina; Tittgemeyer, Marc; Stephan, Klaas E.

    2014-01-01

    Impulsivity plays a key role in decision-making under uncertainty. It is a significant contributor to problem and pathological gambling (PG). Standard assessments of impulsivity by questionnaires, however, have various limitations, partly because impulsivity is a broad, multi-faceted concept. What remains unclear is which of these facets contribute to shaping gambling behavior. In the present study, we investigated impulsivity as expressed in a gambling setting by applying computational modeling to data from 47 healthy male volunteers who played a realistic, virtual slot-machine gambling task. Behaviorally, we found that impulsivity, as measured independently by the 11th revision of the Barratt Impulsiveness Scale (BIS-11), correlated significantly with an aggregate read-out of the following gambling responses: bet increases (BIs), machines switches (MS), casino switches (CS), and double-ups (DUs). Using model comparison, we compared a set of hierarchical Bayesian belief-updating models, i.e., the Hierarchical Gaussian Filter (HGF) and Rescorla–Wagner reinforcement learning (RL) models, with regard to how well they explained different aspects of the behavioral data. We then examined the construct validity of our winning models with multiple regression, relating subject-specific model parameter estimates to the individual BIS-11 total scores. In the most predictive model (a three-level HGF), the two free parameters encoded uncertainty-dependent mechanisms of belief updates and significantly explained BIS-11 variance across subjects. Furthermore, in this model, decision noise was a function of trial-wise uncertainty about winning probability. Collectively, our results provide a proof of concept that hierarchical Bayesian models can characterize the decision-making mechanisms linked to the impulsive traits of an individual. These novel indices of gambling mechanisms unmasked during actual play may be useful for online prevention measures for at-risk players and

  8. LITERATURE ABSTRACTS.

    PubMed

    1971-12-01

    1. General Principles: 'Statistical Aspect of the Correlation Between Objective and Subjective Measurements of Meat Tenderness', by M. C. Gacula, Jr., J. B. Reaume, K. J. Morgan, and R. L. Luckett 1. General Principles: 'Texture of Semi-Solid Foods: Sensory and Physical Correlates', by W. F. Henry, M. H. Katz, F. J. Pilgrim, and A. T. May 2. Instrumentation and Methodology: 'Measurement of Bread Staling', by W. Morandini and L. Wassermann 2. Instrumentation and Methodology: 'Physical Considerations of the Methods of Consistency Measurement of Butter', by E. Knoop 2. Instrumentation and Methodology: 'Electronic Recording Mixers for the Baking Test', by P. W. Voisey, V. M. Bendelow and H. Miller 2. Instrumentation and Methodology: 'Measurement of the Consistency of Reconstituted Instant Potato Flakes', by P. W. Voisey and P. R. Dean 2. Instrumentation and Methodology: 'The Ottawa Electronic Recording Farinograph', by P. W. Voisey, H. Miller and P. L. Byrne 3. Objective Measurements: A. FOODS: 'The Rheological Properties of Corn Horny Endosperm', by J. R. Hamerle*, R. K. White**, and N. N. Mohsenin*** 3. Objective Measurements: 'Evaluation of Mechanical Properties of Comminuted Sausages by Construction and Analysis of Rheological Model', by St. Tyszkiewicz 3. Objective Measurements: 'Studies on Creep Compliance of Butter', by M. Chwiej 3. Objective Measurements: 'Heat-Induced Milk Gels. II. Preparation of Gels and Measurement of Firmness', by M. Kalab, P. W. Voisey and D. B. Emmons 3. Objective Measurements: 'Rheology of Fresh, Aged and Gamma-Irradiated Egg White', by M. A. Tung, J. F. Richards, B. C. Morrison and E. L. Watson 3. Objective Measurements: 'Retardation of Bread Staling - Practical Experiences', by W. Morandini and L. Wassermann 3. Objective Measurements: B. PHARMACEUTICALS: 'Influence of HLB on Certain Physicochemical Parameters of an O/W Emulsion', by M. Schrenzel 3. Objective Measurements: 'The Rheological Evaluation of Semisolids', by L. H. Block and

  9. Quantifying surgical complexity with machine learning: looking beyond patient factors to improve surgical models.

    PubMed

    Van Esbroeck, Alexander; Rubinfeld, Ilan; Hall, Bruce; Syed, Zeeshan

    2014-11-01

    To investigate the use of machine learning to empirically determine the risk of individual surgical procedures and to improve surgical models with this information. American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) data from 2005 to 2009 were used to train support vector machine (SVM) classifiers to learn the relationship between textual constructs in current procedural terminology (CPT) descriptions and mortality, morbidity, Clavien 4 complications, and surgical-site infections (SSI) within 30 days of surgery. The procedural risk scores produced by the SVM classifiers were validated on data from 2010 in univariate and multivariate analyses. The procedural risk scores produced by the SVM classifiers achieved moderate-to-high levels of discrimination in univariate analyses (area under receiver operating characteristic curve: 0.871 for mortality, 0.789 for morbidity, 0.791 for SSI, 0.845 for Clavien 4 complications). Addition of these scores also substantially improved multivariate models comprising patient factors and previously proposed correlates of procedural risk (net reclassification improvement and integrated discrimination improvement: 0.54 and 0.001 for mortality, 0.46 and 0.011 for morbidity, 0.68 and 0.022 for SSI, 0.44 and 0.001 for Clavien 4 complications; P < .05 for all comparisons). Similar improvements were noted in discrimination and calibration for other statistical measures, and in subcohorts comprising patients with general or vascular surgery. Machine learning provides clinically useful estimates of surgical risk for individual procedures. This information can be measured in an entirely data-driven manner and substantially improves multifactorial models to predict postoperative complications. Copyright © 2014 Elsevier Inc. All rights reserved.

  10. A machine learning approach to the potential-field method for implicit modeling of geological structures

    NASA Astrophysics Data System (ADS)

    Gonçalves, Ítalo Gomes; Kumaira, Sissa; Guadagnin, Felipe

    2017-06-01

    Implicit modeling has experienced a rise in popularity over the last decade due to its advantages in terms of speed and reproducibility in comparison with manual digitization of geological structures. The potential-field method consists in interpolating a scalar function that indicates to which side of a geological boundary a given point belongs to, based on cokriging of point data and structural orientations. This work proposes a vector potential-field solution from a machine learning perspective, recasting the problem as multi-class classification, which alleviates some of the original method's assumptions. The potentials related to each geological class are interpreted in a compositional data framework. Variogram modeling is avoided through the use of maximum likelihood to train the model, and an uncertainty measure is introduced. The methodology was applied to the modeling of a sample dataset provided with the software Move™. The calculations were implemented in the R language and 3D visualizations were prepared with the rgl package.

  11. Sugeno-Fuzzy Expert System Modeling for Quality Prediction of Non-Contact Machining Process

    NASA Astrophysics Data System (ADS)

    Sivaraos; Khalim, A. Z.; Salleh, M. S.; Sivakumar, D.; Kadirgama, K.

    2018-03-01

    Modeling can be categorised into four main domains: prediction, optimisation, estimation and calibration. In this paper, the Takagi-Sugeno-Kang (TSK) fuzzy logic method is examined as a prediction modelling method to investigate the taper quality of laser lathing, which seeks to replace traditional lathe machines with 3D laser lathing in order to achieve the desired cylindrical shape of stock materials. Three design parameters were selected: feed rate, cutting speed and depth of cut. A total of twenty-four experiments were conducted with eight sequential runs and replicated three times. The results were found to be 99% of accuracy rate of the TSK fuzzy predictive model, which suggests that the model is a suitable and practical method for non-linear laser lathing process.

  12. Model development, testing and experimentation in a CyberWorkstation for Brain-Machine Interface research.

    PubMed

    Rattanatamrong, Prapaporn; Matsunaga, Andrea; Raiturkar, Pooja; Mesa, Diego; Zhao, Ming; Mahmoudi, Babak; Digiovanna, Jack; Principe, Jose; Figueiredo, Renato; Sanchez, Justin; Fortes, Jose

    2010-01-01

    The CyberWorkstation (CW) is an advanced cyber-infrastructure for Brain-Machine Interface (BMI) research. It allows the development, configuration and execution of BMI computational models using high-performance computing resources. The CW's concept is implemented using a software structure in which an "experiment engine" is used to coordinate all software modules needed to capture, communicate and process brain signals and motor-control commands. A generic BMI-model template, which specifies a common interface to the CW's experiment engine, and a common communication protocol enable easy addition, removal or replacement of models without disrupting system operation. This paper reviews the essential components of the CW and shows how templates can facilitate the processes of BMI model development, testing and incorporation into the CW. It also discusses the ongoing work towards making this process infrastructure independent.

  13. Modeling and prediction of human word search behavior in interactive machine translation

    NASA Astrophysics Data System (ADS)

    Ji, Duo; Yu, Bai; Ma, Bin; Ye, Na

    2017-12-01

    As a kind of computer aided translation method, Interactive Machine Translation technology reduced manual translation repetitive and mechanical operation through a variety of methods, so as to get the translation efficiency, and played an important role in the practical application of the translation work. In this paper, we regarded the behavior of users' frequently searching for words in the translation process as the research object, and transformed the behavior to the translation selection problem under the current translation. The paper presented a prediction model, which is a comprehensive utilization of alignment model, translation model and language model of the searching words behavior. It achieved a highly accurate prediction of searching words behavior, and reduced the switching of mouse and keyboard operations in the users' translation process.

  14. A wearable computing platform for developing cloud-based machine learning models for health monitoring applications.

    PubMed

    Patel, Shyamal; McGinnis, Ryan S; Silva, Ikaro; DiCristofaro, Steve; Mahadevan, Nikhil; Jortberg, Elise; Franco, Jaime; Martin, Albert; Lust, Joseph; Raj, Milan; McGrane, Bryan; DePetrillo, Paolo; Aranyosi, A J; Ceruolo, Melissa; Pindado, Jesus; Ghaffari, Roozbeh

    2016-08-01

    Wearable sensors have the potential to enable clinical-grade ambulatory health monitoring outside the clinic. Technological advances have enabled development of devices that can measure vital signs with great precision and significant progress has been made towards extracting clinically meaningful information from these devices in research studies. However, translating measurement accuracies achieved in the controlled settings such as the lab and clinic to unconstrained environments such as the home remains a challenge. In this paper, we present a novel wearable computing platform for unobtrusive collection of labeled datasets and a new paradigm for continuous development, deployment and evaluation of machine learning models to ensure robust model performance as we transition from the lab to home. Using this system, we train activity classification models across two studies and track changes in model performance as we go from constrained to unconstrained settings.

  15. Study of Two-Dimensional Compressible Non-Acoustic Modeling of Stirling Machine Type Components

    NASA Technical Reports Server (NTRS)

    Tew, Roy C., Jr.; Ibrahim, Mounir B.

    2001-01-01

    A two-dimensional (2-D) computer code was developed for modeling enclosed volumes of gas with oscillating boundaries, such as Stirling machine components. An existing 2-D incompressible flow computer code, CAST, was used as the starting point for the project. CAST was modified to use the compressible non-acoustic Navier-Stokes equations to model an enclosed volume including an oscillating piston. The devices modeled have low Mach numbers and are sufficiently small that the time required for acoustics to propagate across them is negligible. Therefore, acoustics were excluded to enable more time efficient computation. Background information about the project is presented. The compressible non-acoustic flow assumptions are discussed. The governing equations used in the model are presented in transport equation format. A brief description is given of the numerical methods used. Comparisons of code predictions with experimental data are then discussed.

  16. Modelling soil water retention using support vector machines with genetic algorithm optimisation.

    PubMed

    Lamorski, Krzysztof; Sławiński, Cezary; Moreno, Felix; Barna, Gyöngyi; Skierucha, Wojciech; Arrue, José L

    2014-01-01

    This work presents point pedotransfer function (PTF) models of the soil water retention curve. The developed models allowed for estimation of the soil water content for the specified soil water potentials: -0.98, -3.10, -9.81, -31.02, -491.66, and -1554.78 kPa, based on the following soil characteristics: soil granulometric composition, total porosity, and bulk density. Support Vector Machines (SVM) methodology was used for model development. A new methodology for elaboration of retention function models is proposed. Alternative to previous attempts known from literature, the ν-SVM method was used for model development and the results were compared with the formerly used the C-SVM method. For the purpose of models' parameters search, genetic algorithms were used as an optimisation framework. A new form of the aim function used for models parameters search is proposed which allowed for development of models with better prediction capabilities. This new aim function avoids overestimation of models which is typically encountered when root mean squared error is used as an aim function. Elaborated models showed good agreement with measured soil water retention data. Achieved coefficients of determination values were in the range 0.67-0.92. Studies demonstrated usability of ν-SVM methodology together with genetic algorithm optimisation for retention modelling which gave better performing models than other tested approaches.

  17. Discriminative feature-rich models for syntax-based machine translation.

    SciTech Connect

    Dixon, Kevin R.

    This report describes the campus executive LDRD %E2%80%9CDiscriminative Feature-Rich Models for Syntax-Based Machine Translation,%E2%80%9D which was an effort to foster a better relationship between Sandia and Carnegie Mellon University (CMU). The primary purpose of the LDRD was to fund the research of a promising graduate student at CMU; in this case, Kevin Gimpel was selected from the pool of candidates. This report gives a brief overview of Kevin Gimpel's research.

  18. Modeling of Residual Stress and Machining Distortion in Aerospace Components (PREPRINT)

    DTIC Science & Technology

    2010-03-01

    John Gayda, “The Effect of Heat Treatment on Residual Stress and Machining Distortions in Advanced Nickel Base Disk Alloys,” NASA/TM-2001-210717. 2...Wei-Tsu Wu, Guoji Li, Juipeng Tang, Shesh Srivatsa, Ravi Shankar, Ron Wallis, Padu Ramasundaram and John Gayda, “A process modeling system for heat...Materials Processing Technology 98 (2000) 189-195. 6. M.A. Rist, S. Tin, B.A. Roder, J.A. James, and M.R. Daymond , “Residual Stresses in a

  19. Ascertaining Validity in the Abstract Realm of PMESII Simulation Models: An Analysis of the Peace Support Operations Model (PSOM)

    DTIC Science & Technology

    2009-06-01

    simulation is the campaign-level Peace Support Operations Model (PSOM). This thesis provides a quantitative analysis of PSOM. The results are based ...multiple potential outcomes , further development and analysis is required before the model is used for large scale analysis . 15. NUMBER OF PAGES 159...multiple potential outcomes , further development and analysis is required before the model is used for large scale analysis . vi THIS PAGE

  20. Mapping the Transmission Risk of Zika Virus using Machine Learning Models.

    PubMed

    Jiang, Dong; Hao, Mengmeng; Ding, Fangyu; Fu, Jingying; Li, Meng

    2018-06-19

    Zika virus, which has been linked to severe congenital abnormalities, is exacerbating global public health problems with its rapid transnational expansion fueled by increased global travel and trade. Suitability mapping of the transmission risk of Zika virus is essential for drafting public health plans and disease control strategies, which are especially important in areas where medical resources are relatively scarce. Predicting the risk of Zika virus outbreak has been studied in recent years, but the published literature rarely includes multiple model comparisons or predictive uncertainty analysis. Here, three relatively popular machine learning models including backward propagation neural network (BPNN), gradient boosting machine (GBM) and random forest (RF) were adopted to map the probability of Zika epidemic outbreak at the global level, pairing high-dimensional multidisciplinary covariate layers with comprehensive location data on recorded Zika virus infection in humans. The results show that the predicted high-risk areas for Zika transmission are concentrated in four regions: Southeastern North America, Eastern South America, Central Africa and Eastern Asia. To evaluate the performance of machine learning models, the 50 modeling processes were conducted based on a training dataset. The BPNN model obtained the highest predictive accuracy with a 10-fold cross-validation area under the curve (AUC) of 0.966 [95% confidence interval (CI) 0.965-0.967], followed by the GBM model (10-fold cross-validation AUC = 0.964[0.963-0.965]) and the RF model (10-fold cross-validation AUC = 0.963[0.962-0.964]). Based on training samples, compared with the BPNN-based model, we find that significant differences (p = 0.0258* and p = 0.0001***, respectively) are observed for prediction accuracies achieved by the GBM and RF models. Importantly, the prediction uncertainty introduced by the selection of absence data was quantified and could provide more accurate

  1. 3D Magnetic field modeling of a new superconducting synchronous machine using reluctance network method

    NASA Astrophysics Data System (ADS)

    Kelouaz, Moussa; Ouazir, Youcef; Hadjout, Larbi; Mezani, Smail; Lubin, Thiery; Berger, Kévin; Lévêque, Jean

    2018-05-01

    In this paper a new superconducting inductor topology intended for synchronous machine is presented. The studied machine has a standard 3-phase armature and a new kind of 2-poles inductor (claw-pole structure) excited by two coaxial superconducting coils. The air-gap spatial variation of the radial flux density is obtained by inserting a superconducting bulk, which deviates the magnetic field due to the coils. The complex geometry of this inductor usually needs 3D finite elements (FEM) for its analysis. However, to avoid a long computational time inherent to 3D FEM, we propose in this work an alternative modeling, which uses a 3D meshed reluctance network. The results obtained with the developed model are compared to 3D FEM computations as well as to measurements carried out on a laboratory prototype. Finally, a 3D FEM study of the shielding properties of the superconducting screen demonstrates the suitability of using a diamagnetic-like model of the superconducting screen.

  2. Machine learning approaches for estimation of prediction interval for the model output.

    PubMed

    Shrestha, Durga L; Solomatine, Dimitri P

    2006-03-01

    A novel method for estimating prediction uncertainty using machine learning techniques is presented. Uncertainty is expressed in the form of the two quantiles (constituting the prediction interval) of the underlying distribution of prediction errors. The idea is to partition the input space into different zones or clusters having similar model errors using fuzzy c-means clustering. The prediction interval is constructed for each cluster on the basis of empirical distributions of the errors associated with all instances belonging to the cluster under consideration and propagated from each cluster to the examples according to their membership grades in each cluster. Then a regression model is built for in-sample data using computed prediction limits as targets, and finally, this model is applied to estimate the prediction intervals (limits) for out-of-sample data. The method was tested on artificial and real hydrologic data sets using various machine learning techniques. Preliminary results show that the method is superior to other methods estimating the prediction interval. A new method for evaluating performance for estimating prediction interval is proposed as well.

  3. Use of machine learning methods to reduce predictive error of groundwater models.

    PubMed

    Xu, Tianfang; Valocchi, Albert J; Choi, Jaesik; Amir, Eyal

    2014-01-01

    Quantitative analyses of groundwater flow and transport typically rely on a physically-based model, which is inherently subject to error. Errors in model structure, parameter and data lead to both random and systematic error even in the output of a calibrated model. We develop complementary data-driven models (DDMs) to reduce the predictive error of physically-based groundwater models. Two machine learning techniques, the instance-based weighting and support vector regression, are used to build the DDMs. This approach is illustrated using two real-world case studies of the Republican River Compact Administration model and the Spokane Valley-Rathdrum Prairie model. The two groundwater models have different hydrogeologic settings, parameterization, and calibration methods. In the first case study, cluster analysis is introduced for data preprocessing to make the DDMs more robust and computationally efficient. The DDMs reduce the root-mean-square error (RMSE) of the temporal, spatial, and spatiotemporal prediction of piezometric head of the groundwater model by 82%, 60%, and 48%, respectively. In the second case study, the DDMs reduce the RMSE of the temporal prediction of piezometric head of the groundwater model by 77%. It is further demonstrated that the effectiveness of the DDMs depends on the existence and extent of the structure in the error of the physically-based model. © 2013, National GroundWater Association.

  4. Quality prediction modeling for sintered ores based on mechanism models of sintering and extreme learning machine based error compensation

    NASA Astrophysics Data System (ADS)

    Tiebin, Wu; Yunlian, Liu; Xinjun, Li; Yi, Yu; Bin, Zhang

    2018-06-01

    Aiming at the difficulty in quality prediction of sintered ores, a hybrid prediction model is established based on mechanism models of sintering and time-weighted error compensation on the basis of the extreme learning machine (ELM). At first, mechanism models of drum index, total iron, and alkalinity are constructed according to the chemical reaction mechanism and conservation of matter in the sintering process. As the process is simplified in the mechanism models, these models are not able to describe high nonlinearity. Therefore, errors are inevitable. For this reason, the time-weighted ELM based error compensation model is established. Simulation results verify that the hybrid model has a high accuracy and can meet the requirement for industrial applications.

  5. Hybrid wavelet-support vector machine approach for modelling rainfall-runoff process.

    PubMed

    Komasi, Mehdi; Sharghi, Soroush

    2016-01-01

    Because of the importance of water resources management, the need for accurate modeling of the rainfall-runoff process has rapidly grown in the past decades. Recently, the support vector machine (SVM) approach has been used by hydrologists for rainfall-runoff modeling and the other fields of hydrology. Similar to the other artificial intelligence models, such as artificial neural network (ANN) and adaptive neural fuzzy inference system, the SVM model is based on the autoregressive properties. In this paper, the wavelet analysis was linked to the SVM model concept for modeling the rainfall-runoff process of Aghchai and Eel River watersheds. In this way, the main time series of two variables, rainfall and runoff, were decomposed to multiple frequent time series by wavelet theory; then, these time series were imposed as input data on the SVM model in order to predict the runoff discharge one day ahead. The obtained results show that the wavelet SVM model can predict both short- and long-term runoff discharges by considering the seasonality effects. Also, the proposed hybrid model is relatively more appropriate than classical autoregressive ones such as ANN and SVM because it uses the multi-scale time series of rainfall and runoff data in the modeling process.

  6. Glucose Oxidase Biosensor Modeling and Predictors Optimization by Machine Learning Methods.

    PubMed

    Gonzalez-Navarro, Felix F; Stilianova-Stoytcheva, Margarita; Renteria-Gutierrez, Livier; Belanche-Muñoz, Lluís A; Flores-Rios, Brenda L; Ibarra-Esquer, Jorge E

    2016-10-26

    Biosensors are small analytical devices incorporating a biological recognition element and a physico-chemical transducer to convert a biological signal into an electrical reading. Nowadays, their technological appeal resides in their fast performance, high sensitivity and continuous measuring capabilities; however, a full understanding is still under research. This paper aims to contribute to this growing field of biotechnology, with a focus on Glucose-Oxidase Biosensor (GOB) modeling through statistical learning methods from a regression perspective. We model the amperometric response of a GOB with dependent variables under different conditions, such as temperature, benzoquinone, pH and glucose concentrations, by means of several machine learning algorithms. Since the sensitivity of a GOB response is strongly related to these dependent variables, their interactions should be optimized to maximize the output signal, for which a genetic algorithm and simulated annealing are used. We report a model that shows a good generalization error and is consistent with the optimization.

  7. Extreme learning machine for reduced order modeling of turbulent geophysical flows.

    PubMed

    San, Omer; Maulik, Romit

    2018-04-01

    We investigate the application of artificial neural networks to stabilize proper orthogonal decomposition-based reduced order models for quasistationary geophysical turbulent flows. An extreme learning machine concept is introduced for computing an eddy-viscosity closure dynamically to incorporate the effects of the truncated modes. We consider a four-gyre wind-driven ocean circulation problem as our prototype setting to assess the performance of the proposed data-driven approach. Our framework provides a significant reduction in computational time and effectively retains the dynamics of the full-order model during the forward simulation period beyond the training data set. Furthermore, we show that the method is robust for larger choices of time steps and can be used as an efficient and reliable tool for long time integration of general circulation models.

  8. Glucose Oxidase Biosensor Modeling and Predictors Optimization by Machine Learning Methods †

    PubMed Central

    Gonzalez-Navarro, Felix F.; Stilianova-Stoytcheva, Margarita; Renteria-Gutierrez, Livier; Belanche-Muñoz, Lluís A.; Flores-Rios, Brenda L.; Ibarra-Esquer, Jorge E.

    2016-01-01

    Biosensors are small analytical devices incorporating a biological recognition element and a physico-chemical transducer to convert a biological signal into an electrical reading. Nowadays, their technological appeal resides in their fast performance, high sensitivity and continuous measuring capabilities; however, a full understanding is still under research. This paper aims to contribute to this growing field of biotechnology, with a focus on Glucose-Oxidase Biosensor (GOB) modeling through statistical learning methods from a regression perspective. We model the amperometric response of a GOB with dependent variables under different conditions, such as temperature, benzoquinone, pH and glucose concentrations, by means of several machine learning algorithms. Since the sensitivity of a GOB response is strongly related to these dependent variables, their interactions should be optimized to maximize the output signal, for which a genetic algorithm and simulated annealing are used. We report a model that shows a good generalization error and is consistent with the optimization. PMID:27792165

  9. Specific modes of vibratory technological machines: mathematical models, peculiarities of interaction of system elements

    NASA Astrophysics Data System (ADS)

    Eliseev, A. V.; Sitov, I. S.; Eliseev, S. V.

    2018-03-01

    The methodological basis of constructing mathematical models of vibratory technological machines is developed in the article. An approach is proposed that makes it possible to introduce a vibration table in a specific mode that provides conditions for the dynamic damping of oscillations for the zone of placement of a vibration exciter while providing specified vibration parameters in the working zone of the vibration table. The aim of the work is to develop methods of mathematical modeling, oriented to technological processes with long cycles. The technologies of structural mathematical modeling are used with structural schemes, transfer functions and amplitude-frequency characteristics. The concept of the work is to test the possibilities of combining the conditions for reducing loads with working components of a vibration exciter while simultaneously maintaining sufficiently wide limits in variating the parameters of the vibrational field.

  10. Extreme learning machine for reduced order modeling of turbulent geophysical flows

    NASA Astrophysics Data System (ADS)

    San, Omer; Maulik, Romit

    2018-04-01

    We investigate the application of artificial neural networks to stabilize proper orthogonal decomposition-based reduced order models for quasistationary geophysical turbulent flows. An extreme learning machine concept is introduced for computing an eddy-viscosity closure dynamically to incorporate the effects of the truncated modes. We consider a four-gyre wind-driven ocean circulation problem as our prototype setting to assess the performance of the proposed data-driven approach. Our framework provides a significant reduction in computational time and effectively retains the dynamics of the full-order model during the forward simulation period beyond the training data set. Furthermore, we show that the method is robust for larger choices of time steps and can be used as an efficient and reliable tool for long time integration of general circulation models.

  11. MIP models and hybrid algorithms for simultaneous job splitting and scheduling on unrelated parallel machines.

    PubMed

    Eroglu, Duygu Yilmaz; Ozmutlu, H Cenk

    2014-01-01

    We developed mixed integer programming (MIP) models and hybrid genetic-local search algorithms for the scheduling problem of unrelated parallel machines with job sequence and machine-dependent setup times and with job splitting property. The first contribution of this paper is to introduce novel algorithms which make splitting and scheduling simultaneously with variable number of subjobs. We proposed simple chromosome structure which is constituted by random key numbers in hybrid genetic-local search algorithm (GAspLA). Random key numbers are used frequently in genetic algorithms, but it creates additional difficulty when hybrid factors in local search are implemented. We developed algorithms that satisfy the adaptation of results of local search into the genetic algorithms with minimum relocation operation of genes' random key numbers. This is the second contribution of the paper. The third contribution of this paper is three developed new MIP models which are making splitting and scheduling simultaneously. The fourth contribution of this paper is implementation of the GAspLAMIP. This implementation let us verify the optimality of GAspLA for the studied combinations. The proposed methods are tested on a set of problems taken from the literature and the results validate the effectiveness of the proposed algorithms.

  12. MIP Models and Hybrid Algorithms for Simultaneous Job Splitting and Scheduling on Unrelated Parallel Machines

    PubMed Central

    Ozmutlu, H. Cenk

    2014-01-01

    We developed mixed integer programming (MIP) models and hybrid genetic-local search algorithms for the scheduling problem of unrelated parallel machines with job sequence and machine-dependent setup times and with job splitting property. The first contribution of this paper is to introduce novel algorithms which make splitting and scheduling simultaneously with variable number of subjobs. We proposed simple chromosome structure which is constituted by random key numbers in hybrid genetic-local search algorithm (GAspLA). Random key numbers are used frequently in genetic algorithms, but it creates additional difficulty when hybrid factors in local search are implemented. We developed algorithms that satisfy the adaptation of results of local search into the genetic algorithms with minimum relocation operation of genes' random key numbers. This is the second contribution of the paper. The third contribution of this paper is three developed new MIP models which are making splitting and scheduling simultaneously. The fourth contribution of this paper is implementation of the GAspLAMIP. This implementation let us verify the optimality of GAspLA for the studied combinations. The proposed methods are tested on a set of problems taken from the literature and the results validate the effectiveness of the proposed algorithms. PMID:24977204

  13. Development of machine learning models to predict inhibition of 3-dehydroquinate dehydratase.

    PubMed

    de Ávila, Maurício Boff; de Azevedo, Walter Filgueira

    2018-04-20

    In this study, we describe the development of new machine learning models to predict inhibition of the enzyme 3-dehydroquinate dehydratase (DHQD). This enzyme is the third step of the shikimate pathway and is responsible for the synthesis of chorismate, which is a natural precursor of aromatic amino acids. The enzymes of shikimate pathway are absent in humans, which make them protein targets for the design of antimicrobial drugs. We focus our study on the crystallographic structures of DHQD in complex with competitive inhibitors, for which experimental inhibition constant data is available. Application of supervised machine learning techniques was able to elaborate a robust DHQD-targeted model to predict binding affinity. Combination of high-resolution crystallographic structures and binding information indicates that the prevalence of intermolecular electrostatic interactions between DHQD and competitive inhibitors is of pivotal importance for the binding affinity against this enzyme. The present findings can be used to speed up virtual screening studies focused on the DHQD structure. © 2018 John Wiley & Sons A/S.

  14. A hybrid flowshop scheduling model considering dedicated machines and lot-splitting for the solar cell industry

    NASA Astrophysics Data System (ADS)

    Wang, Li-Chih; Chen, Yin-Yann; Chen, Tzu-Li; Cheng, Chen-Yang; Chang, Chin-Wei

    2014-10-01

    This paper studies a solar cell industry scheduling problem, which is similar to traditional hybrid flowshop scheduling (HFS). In a typical HFS problem, the allocation of machine resources for each order should be scheduled in advance. However, the challenge in solar cell manufacturing is the number of machines that can be adjusted dynamically to complete the job. An optimal production scheduling model is developed to explore these issues, considering the practical characteristics, such as hybrid flowshop, parallel machine system, dedicated machines, sequence independent job setup times and sequence dependent job setup times. The objective of this model is to minimise the makespan and to decide the processing sequence of the orders/lots in each stage, lot-splitting decisions for the orders and the number of machines used to satisfy the demands in each stage. From the experimental results, lot-splitting has significant effect on shortening the makespan, and the improvement effect is influenced by the processing time and the setup time of orders. Therefore, the threshold point to improve the makespan can be identified. In addition, the model also indicates that more lot-splitting approaches, that is, the flexibility of allocating orders/lots to machines is larger, will result in a better scheduling performance.

  15. State Event Models for the Formal Analysis of Human-Machine Interactions

    NASA Technical Reports Server (NTRS)

    Combefis, Sebastien; Giannakopoulou, Dimitra; Pecheur, Charles

    2014-01-01

    The work described in this paper was motivated by our experience with applying a framework for formal analysis of human-machine interactions (HMI) to a realistic model of an autopilot. The framework is built around a formally defined conformance relation called "fullcontrol" between an actual system and the mental model according to which the system is operated. Systems are well-designed if they can be described by relatively simple, full-control, mental models for their human operators. For this reason, our framework supports automated generation of minimal full-control mental models for HMI systems, where both the system and the mental models are described as labelled transition systems (LTS). The autopilot that we analysed has been developed in the NASA Ames HMI prototyping tool ADEPT. In this paper, we describe how we extended the models that our HMI analysis framework handles to allow adequate representation of ADEPT models. We then provide a property-preserving reduction from these extended models to LTSs, to enable application of our LTS-based formal analysis algorithms. Finally, we briefly discuss the analyses we were able to perform on the autopilot model with our extended framework.

  16. Cheminformatic models based on machine learning for pyruvate kinase inhibitors of Leishmania mexicana.

    PubMed

    Jamal, Salma; Scaria, Vinod

    2013-11-19

    Leishmaniasis is a neglected tropical disease which affects approx. 12 million individuals worldwide and caused by parasite Leishmania. The current drugs used in the treatment of Leishmaniasis are highly toxic and has seen widespread emergence of drug resistant strains which necessitates the need for the development of new therapeutic options. The high throughput screen data available has made it possible to generate computational predictive models which have the ability to assess the active scaffolds in a chemical library followed by its ADME/toxicity properties in the biological trials. In the present study, we have used publicly available, high-throughput screen datasets of chemical moieties which have been adjudged to target the pyruvate kinase enzyme of L. mexicana (LmPK). The machine learning approach was used to create computational models capable of predicting the biological activity of novel antileishmanial compounds. Further, we evaluated the molecules using the substructure based approach to identify the common substructures contributing to their activity. We generated computational models based on machine learning methods and evaluated the performance of these models based on various statistical figures of merit. Random forest based approach was determined to be the most sensitive, better accuracy as well as ROC. We further added a substructure based approach to analyze the molecules to identify potentially enriched substructures in the active dataset. We believe that the models developed in the present study would lead to reduction in cost and length of clinical studies and hence newer drugs would appear faster in the market providing better healthcare options to the patients.

  17. Modeling Dengue vector population using remotely sensed data and machine learning.

    PubMed

    Scavuzzo, Juan M; Trucco, Francisco; Espinosa, Manuel; Tauro, Carolina B; Abril, Marcelo; Scavuzzo, Carlos M; Frery, Alejandro C

    2018-05-16

    Mosquitoes are vectors of many human diseases. In particular, Aedes ægypti (Linnaeus) is the main vector for Chikungunya, Dengue, and Zika viruses in Latin America and it represents a global threat. Public health policies that aim at combating this vector require dependable and timely information, which is usually expensive to obtain with field campaigns. For this reason, several efforts have been done to use remote sensing due to its reduced cost. The present work includes the temporal modeling of the oviposition activity (measured weekly on 50 ovitraps in a north Argentinean city) of Aedes ægypti (Linnaeus), based on time series of data extracted from operational earth observation satellite images. We use are NDVI, NDWI, LST night, LST day and TRMM-GPM rain from 2012 to 2016 as predictive variables. In contrast to previous works which use linear models, we employ Machine Learning techniques using completely accessible open source toolkits. These models have the advantages of being non-parametric and capable of describing nonlinear relationships between variables. Specifically, in addition to two linear approaches, we assess a support vector machine, an artificial neural networks, a K-nearest neighbors and a decision tree regressor. Considerations are made on parameter tuning and the validation and training approach. The results are compared to linear models used in previous works with similar data sets for generating temporal predictive models. These new tools perform better than linear approaches, in particular nearest neighbor regression (KNNR) performs the best. These results provide better alternatives to be implemented operatively on the Argentine geospatial risk system that is running since 2012. Copyright © 2018 Elsevier B.V. All rights reserved.

  18. Subgrid-scale scalar flux modelling based on optimal estimation theory and machine-learning procedures

    NASA Astrophysics Data System (ADS)

    Vollant, A.; Balarac, G.; Corre, C.

    2017-09-01

    New procedures are explored for the development of models in the context of large eddy simulation (LES) of a passive scalar. They rely on the combination of the optimal estimator theory with machine-learning algorithms. The concept of optimal estimator allows to identify the most accurate set of parameters to be used when deriving a model. The model itself can then be defined by training an artificial neural network (ANN) on a database derived from the filtering of direct numerical simulation (DNS) results. This procedure leads to a subgrid scale model displaying good structural performance, which allows to perform LESs very close to the filtered DNS results. However, this first procedure does not control the functional performance so that the model can fail when the flow configuration differs from the training database. Another procedure is then proposed, where the model functional form is imposed and the ANN used only to define the model coefficients. The training step is a bi-objective optimisation in order to control both structural and functional performances. The model derived from this second procedure proves to be more robust. It also provides stable LESs for a turbulent plane jet flow configuration very far from the training database but over-estimates the mixing process in that case.

  19. Deriving global parameter estimates for the Noah land surface model using FLUXNET and machine learning

    NASA Astrophysics Data System (ADS)

    Chaney, Nathaniel W.; Herman, Jonathan D.; Ek, Michael B.; Wood, Eric F.

    2016-11-01

    With their origins in numerical weather prediction and climate modeling, land surface models aim to accurately partition the surface energy balance. An overlooked challenge in these schemes is the role of model parameter uncertainty, particularly at unmonitored sites. This study provides global parameter estimates for the Noah land surface model using 85 eddy covariance sites in the global FLUXNET network. The at-site parameters are first calibrated using a Latin Hypercube-based ensemble of the most sensitive parameters, determined by the Sobol method, to be the minimum stomatal resistance (rs,min), the Zilitinkevich empirical constant (Czil), and the bare soil evaporation exponent (fxexp). Calibration leads to an increase in the mean Kling-Gupta Efficiency performance metric from 0.54 to 0.71. These calibrated parameter sets are then related to local environmental characteristics using the Extra-Trees machine learning algorithm. The fitted Extra-Trees model is used to map the optimal parameter sets over the globe at a 5 km spatial resolution. The leave-one-out cross validation of the mapped parameters using the Noah land surface model suggests that there is the potential to skillfully relate calibrated model parameter sets to local environmental characteristics. The results demonstrate the potential to use FLUXNET to tune the parameterizations of surface fluxes in land surface models and to provide improved parameter estimates over the globe.

  20. Uncertainty analysis of an inflow forecasting model: extension of the UNEEC machine learning-based method

    NASA Astrophysics Data System (ADS)

    Pianosi, Francesca; Lal Shrestha, Durga; Solomatine, Dimitri

    2010-05-01

    This research presents an extension of UNEEC (Uncertainty Estimation based on Local Errors and Clustering, Shrestha and Solomatine, 2006, 2008 & Solomatine and Shrestha, 2009) method in the direction of explicit inclusion of parameter uncertainty. UNEEC method assumes that there is an optimal model and the residuals of the model can be used to assess the uncertainty of the model prediction. It is assumed that all sources of uncertainty including input, parameter and model structure uncertainty are explicitly manifested in the model residuals. In this research, theses assumptions are relaxed, and the UNEEC method is extended to consider parameter uncertainty as well (abbreviated as UNEEC-P). In UNEEC-P, first we use Monte Carlo (MC) sampling in parameter space to generate N model realizations (each of which is a time series), estimate the prediction quantiles based on the empirical distribution functions of the model residuals considering all the residual realizations, and only then apply the standard UNEEC method that encapsulates the uncertainty of a hydrologic model (expressed by quantiles of the error distribution) in a machine learning model (e.g., ANN). UNEEC-P is applied first to a linear regression model of synthetic data, and then to a real case study of forecasting inflow to lake Lugano in northern Italy. The inflow forecasting model is a stochastic heteroscedastic model (Pianosi and Soncini-Sessa, 2009). The preliminary results show that the UNEEC-P method produces wider uncertainty bounds, which is consistent with the fact that the method considers also parameter uncertainty of the optimal model. In the future UNEEC method will be further extended to consider input and structure uncertainty which will provide more realistic estimation of model predictions.

  1. Using fuzzy models in machining control system and assessment of sustainability

    NASA Astrophysics Data System (ADS)

    Grinek, A. V.; Boychuk, I. P.; Dantsevich, I. M.

    2018-03-01

    Description of the complex relationship of the optimum velocity with the temperature-strength state in the cutting zone for machining a fuzzy model is proposed. The fuzzy-logical conclusion allows determining the processing speed, which ensures effective, from the point of view of ensuring the quality of the surface layer, the temperature in the cutting zone and the maximum allowable cutting force. A scheme for stabilizing the temperature-strength state in the cutting zone using a nonlinear fuzzy PD–controller is proposed. The stability of the nonlinear system is estimated with the help of grapho–analytical realization of the method of harmonic balance and by modeling in MatLab.

  2. Quick Estimation Model for the Concentration of Indoor Airborne Culturable Bacteria: An Application of Machine Learning.

    PubMed

    Liu, Zhijian; Li, Hao; Cao, Guoqing

    2017-07-30

    Indoor airborne culturable bacteria are sometimes harmful to human health. Therefore, a quick estimation of their concentration is particularly necessary. However, measuring the indoor microorganism concentration (e.g., bacteria) usually requires a large amount of time, economic cost, and manpower. In this paper, we aim to provide a quick solution: using knowledge-based machine learning to provide quick estimation of the concentration of indoor airborne culturable bacteria only with the inputs of several measurable indoor environmental indicators, including: indoor particulate matter (PM 2.5 and PM 10 ), temperature, relative humidity, and CO₂ concentration. Our results show that a general regression neural network (GRNN) model can sufficiently provide a quick and decent estimation based on the model training and testing using an experimental database with 249 data groups.

  3. Software model of a machine vision system based on the common house fly.

    PubMed

    Madsen, Robert; Barrett, Steven; Wilcox, Michael

    2005-01-01

    The vision system of the common house fly has many properties, such as hyperacuity and parallel structure, which would be advantageous in a machine vision system. A software model has been developed which is ultimately intended to be a tool to guide the design of an analog real time vision system. The model starts by laying out cartridges over an image. The cartridges are analogous to the ommatidium of the fly's eye and contain seven photoreceptors each with a Gaussian profile. The spacing between photoreceptors is variable providing for more or less detail as needed. The cartridges provide information on what type of features they see and neighboring cartridges share information to construct a feature map.

  4. Quick Estimation Model for the Concentration of Indoor Airborne Culturable Bacteria: An Application of Machine Learning

    PubMed Central

    Liu, Zhijian; Li, Hao; Cao, Guoqing

    2017-01-01

    Indoor airborne culturable bacteria are sometimes harmful to human health. Therefore, a quick estimation of their concentration is particularly necessary. However, measuring the indoor microorganism concentration (e.g., bacteria) usually requires a large amount of time, economic cost, and manpower. In this paper, we aim to provide a quick solution: using knowledge-based machine learning to provide quick estimation of the concentration of indoor airborne culturable bacteria only with the inputs of several measurable indoor environmental indicators, including: indoor particulate matter (PM2.5 and PM10), temperature, relative humidity, and CO2 concentration. Our results show that a general regression neural network (GRNN) model can sufficiently provide a quick and decent estimation based on the model training and testing using an experimental database with 249 data groups. PMID:28758941

  5. 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

  6. 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.

  7. Machine learning algorithms for modeling groundwater level changes in agricultural regions of the U.S.

    SciTech Connect

    Sahoo, S.; Russo, T. A.; Elliott, J.

    Climate, groundwater extraction, and surface water flows have complex nonlinear relationships with groundwater level in agricultural regions. To better understand the relative importance of each driver and predict groundwater level change, we develop a new ensemble modeling framework based on spectral analysis, machine learning, and uncertainty analysis, as an alternative to complex and computationally expensive physical models. We apply and evaluate this new approach in the context of two aquifer systems supporting agricultural production in the United States: the High Plains aquifer (HPA) and the Mississippi River Valley alluvial aquifer (MRVA). We select input data sets by using a combinationmore » of mutual information, genetic algorithms, and lag analysis, and then use the selected data sets in a Multilayer Perceptron network architecture to simulate seasonal groundwater level change. As expected, model results suggest that irrigation demand has the highest influence on groundwater level change for a majority of the wells. The subset of groundwater observations not used in model training or cross-validation correlates strongly (R > 0.8) with model results for 88 and 83% of the wells in the HPA and MRVA, respectively. In both aquifer systems, the error in the modeled cumulative groundwater level change during testing (2003-2012) was less than 2 m over a majority of the area. Here, we conclude that our modeling framework can serve as an alternative approach to simulating groundwater level change and water availability, especially in regions where subsurface properties are unknown.« less

  8. Machine learning algorithms for modeling groundwater level changes in agricultural regions of the U.S.

    DOE PAGES

    Sahoo, S.; Russo, T. A.; Elliott, J.; ...

    2017-05-13

    Climate, groundwater extraction, and surface water flows have complex nonlinear relationships with groundwater level in agricultural regions. To better understand the relative importance of each driver and predict groundwater level change, we develop a new ensemble modeling framework based on spectral analysis, machine learning, and uncertainty analysis, as an alternative to complex and computationally expensive physical models. We apply and evaluate this new approach in the context of two aquifer systems supporting agricultural production in the United States: the High Plains aquifer (HPA) and the Mississippi River Valley alluvial aquifer (MRVA). We select input data sets by using a combinationmore » of mutual information, genetic algorithms, and lag analysis, and then use the selected data sets in a Multilayer Perceptron network architecture to simulate seasonal groundwater level change. As expected, model results suggest that irrigation demand has the highest influence on groundwater level change for a majority of the wells. The subset of groundwater observations not used in model training or cross-validation correlates strongly (R > 0.8) with model results for 88 and 83% of the wells in the HPA and MRVA, respectively. In both aquifer systems, the error in the modeled cumulative groundwater level change during testing (2003-2012) was less than 2 m over a majority of the area. Here, we conclude that our modeling framework can serve as an alternative approach to simulating groundwater level change and water availability, especially in regions where subsurface properties are unknown.« less

  9. One- and two-dimensional Stirling machine simulation using experimentally generated flow turbulence models

    NASA Technical Reports Server (NTRS)

    Goldberg, Louis F.

    1990-01-01

    Investigations of one- and two-dimensional (1- or 2-D) simulations of Stirling machines centered around experimental data generated by the U. of Minnesota Mechanical Engineering Test Rig (METR) are covered. This rig was used to investigate oscillating flows about a zero mean with emphasis on laminar/turbulent flow transitions in tubes. The Space Power Demonstrator Engine (SPDE) and in particular, its heater, were the subjects of the simulations. The heater was treated as a 1- or 2-D entity in an otherwise 1-D system. The 2-D flow effects impacted the transient flow predictions in the heater itself but did not have a major impact on overall system performance. Information propagation effects may be a significant issue in the simulation (if not the performance) of high-frequency, high-pressure Stirling machines. This was investigated further by comparing a simulation against an experimentally validated analytic solution for the fluid dynamics of a transmission line. The applicability of the pressure-linking algorithm for compressible flows may be limited by characteristic number (defined as flow path information traverses per cycle); this warrants further study. Lastly the METR was simulated in 1- and 2-D. A two-parameter k-w foldback function turbulence model was developed and tested against a limited set of METR experimental data.

  10. A machine learning approach for automated assessment of retinal vasculature in the oxygen induced retinopathy model.

    PubMed

    Mazzaferri, Javier; Larrivée, Bruno; Cakir, Bertan; Sapieha, Przemyslaw; Costantino, Santiago

    2018-03-02

    Preclinical studies of vascular retinal diseases rely on the assessment of developmental dystrophies in the oxygen induced retinopathy rodent model. The quantification of vessel tufts and avascular regions is typically computed manually from flat mounted retinas imaged using fluorescent probes that highlight the vascular network. Such manual measurements are time-consuming and hampered by user variability and bias, thus a rapid and objective method is needed. Here, we introduce a machine learning approach to segment and characterize vascular tufts, delineate the whole vasculature network, and identify and analyze avascular regions. Our quantitative retinal vascular assessment (QuRVA) technique uses a simple machine learning method and morphological analysis to provide reliable computations of vascular density and pathological vascular tuft regions, devoid of user intervention within seconds. We demonstrate the high degree of error and variability of manual segmentations, and designed, coded, and implemented a set of algorithms to perform this task in a fully automated manner. We benchmark and validate the results of our analysis pipeline using the consensus of several manually curated segmentations using commonly used computer tools. The source code of our implementation is released under version 3 of the GNU General Public License ( https://www.mathworks.com/matlabcentral/fileexchange/65699-javimazzaf-qurva ).

  11. Semi-supervised Machine Learning for Analysis of Hydrogeochemical Data and Models

    NASA Astrophysics Data System (ADS)

    Vesselinov, Velimir; O'Malley, Daniel; Alexandrov, Boian; Moore, Bryan

    2017-04-01

    Data- and model-based analyses such as uncertainty quantification, sensitivity analysis, and decision support using complex physics models with numerous model parameters and typically require a huge number of model evaluations (on order of 10^6). Furthermore, model simulations of complex physics may require substantial computational time. For example, accounting for simultaneously occurring physical processes such as fluid flow and biogeochemical reactions in heterogeneous porous medium may require several hours of wall-clock computational time. To address these issues, we have developed a novel methodology for semi-supervised machine learning based on Non-negative Matrix Factorization (NMF) coupled with customized k-means clustering. The algorithm allows for automated, robust Blind Source Separation (BSS) of groundwater types (contamination sources) based on model-free analyses of observed hydrogeochemical data. We have also developed reduced order modeling tools, which coupling support vector regression (SVR), genetic algorithms (GA) and artificial and convolutional neural network (ANN/CNN). SVR is applied to predict the model behavior within prior uncertainty ranges associated with the model parameters. ANN and CNN procedures are applied to upscale heterogeneity of the porous medium. In the upscaling process, fine-scale high-resolution models of heterogeneity are applied to inform coarse-resolution models which have improved computational efficiency while capturing the impact of fine-scale effects at the course scale of interest. These techniques are tested independently on a series of synthetic problems. We also present a decision analysis related to contaminant remediation where the developed reduced order models are applied to reproduce groundwater flow and contaminant transport in a synthetic heterogeneous aquifer. The tools are coded in Julia and are a part of the MADS high-performance computational framework (https://github.com/madsjulia/Mads.jl).

  12. Ecophysiological Modeling of Grapevine Water Stress in Burgundy Terroirs by a Machine-Learning Approach.

    PubMed

    Brillante, Luca; Mathieu, Olivier; Lévêque, Jean; Bois, Benjamin

    2016-01-01

    In a climate change scenario, successful modeling of the relationships between plant-soil-meteorology is crucial for a sustainable agricultural production, especially for perennial crops. Grapevines (Vitis vinifera L. cv Chardonnay) located in eight experimental plots (Burgundy, France) along a hillslope were monitored weekly for 3 years for leaf water potentials, both at predawn (Ψpd) and at midday (Ψstem). The water stress experienced by grapevine was modeled as a function of meteorological data (minimum and maximum temperature, rainfall) and soil characteristics (soil texture, gravel content, slope) by a gradient boosting machine. Model performance was assessed by comparison with carbon isotope discrimination (δ(13)C) of grape sugars at harvest and by the use of a test-set. The developed models reached outstanding prediction performance (RMSE < 0.08 MPa for Ψstem and < 0.06 MPa for Ψpd), comparable to measurement accuracy. Model predictions at a daily time step improved correlation with δ(13)C data, respect to the observed trend at a weekly time scale. The role of each predictor in these models was described in order to understand how temperature, rainfall, soil texture, gravel content and slope affect the grapevine water status in the studied context. This work proposes a straight-forward strategy to simulate plant water stress in field condition, at a local scale; to investigate ecological relationships in the vineyard and adapt cultural practices to future conditions.

  13. Rotary ultrasonic machining of CFRP: a mechanistic predictive model for cutting force.

    PubMed

    Cong, W L; Pei, Z J; Sun, X; Zhang, C L

    2014-02-01

    Cutting force is one of the most important output variables in rotary ultrasonic machining (RUM) of carbon fiber reinforced plastic (CFRP) composites. Many experimental investigations on cutting force in RUM of CFRP have been reported. However, in the literature, there are no cutting force models for RUM of CFRP. This paper develops a mechanistic predictive model for cutting force in RUM of CFRP. The material removal mechanism of CFRP in RUM has been analyzed first. The model is based on the assumption that brittle fracture is the dominant mode of material removal. CFRP micromechanical analysis has been conducted to represent CFRP as an equivalent homogeneous material to obtain the mechanical properties of CFRP from its components. Based on this model, relationships between input variables (including ultrasonic vibration amplitude, tool rotation speed, feedrate, abrasive size, and abrasive concentration) and cutting force can be predicted. The relationships between input variables and important intermediate variables (indentation depth, effective contact time, and maximum impact force of single abrasive grain) have been investigated to explain predicted trends of cutting force. Experiments are conducted to verify the model, and experimental results agree well with predicted trends from this model. Copyright © 2013 Elsevier B.V. All rights reserved.

  14. Hidden Markov models and other machine learning approaches in computational molecular biology

    SciTech Connect

    Baldi, P.

    1995-12-31

    This tutorial was one of eight tutorials selected to be presented at the Third International Conference on Intelligent Systems for Molecular Biology which was held in the United Kingdom from July 16 to 19, 1995. Computational tools are increasingly needed to process the massive amounts of data, to organize and classify sequences, to detect weak similarities, to separate coding from non-coding regions, and reconstruct the underlying evolutionary history. The fundamental problem in machine learning is the same as in scientific reasoning in general, as well as statistical modeling: to come up with a good model for the data. In thismore » tutorial four classes of models are reviewed. They are: Hidden Markov models; artificial Neural Networks; Belief Networks; and Stochastic Grammars. When dealing with DNA and protein primary sequences, Hidden Markov models are one of the most flexible and powerful alignments and data base searches. In this tutorial, attention is focused on the theory of Hidden Markov Models, and how to apply them to problems in molecular biology.« less

  15. Machine Learning-based discovery of closures for reduced models of dynamical systems

    NASA Astrophysics Data System (ADS)

    Pan, Shaowu; Duraisamy, Karthik

    2017-11-01

    Despite the successful application of machine learning (ML) in fields such as image processing and speech recognition, only a few attempts has been made toward employing ML to represent the dynamics of complex physical systems. Previous attempts mostly focus on parameter calibration or data-driven augmentation of existing models. In this work we present a ML framework to discover closure terms in reduced models of dynamical systems and provide insights into potential problems associated with data-driven modeling. Based on exact closure models for linear system, we propose a general linear closure framework from viewpoint of optimization. The framework is based on trapezoidal approximation of convolution term. Hyperparameters that need to be determined include temporal length of memory effect, number of sampling points, and dimensions of hidden states. To circumvent the explicit specification of memory effect, a general framework inspired from neural networks is also proposed. We conduct both a priori and posteriori evaluations of the resulting model on a number of non-linear dynamical systems. This work was supported in part by AFOSR under the project ``LES Modeling of Non-local effects using Statistical Coarse-graining'' with Dr. Jean-Luc Cambier as the technical monitor.

  16. Chemical Kinetics of Hydrogen Atom Abstraction from Allylic Sites by 3O2; Implications for Combustion Modeling and Simulation.

    PubMed

    Zhou, Chong-Wen; Simmie, John M; Somers, Kieran P; Goldsmith, C Franklin; Curran, Henry J

    2017-03-09

    Hydrogen atom abstraction from allylic C-H bonds by molecular oxygen plays a very important role in determining the reactivity of fuel molecules having allylic hydrogen atoms. Rate constants for hydrogen atom abstraction by molecular oxygen from molecules with allylic sites have been calculated. A series of molecules with primary, secondary, tertiary, and super secondary allylic hydrogen atoms of alkene, furan, and alkylbenzene families are taken into consideration. Those molecules include propene, 2-butene, isobutene, 2-methylfuran, and toluene containing the primary allylic hydrogen atom; 1-butene, 1-pentene, 2-ethylfuran, ethylbenzene, and n-propylbenzene containing the secondary allylic hydrogen atom; 3-methyl-1-butene, 2-isopropylfuran, and isopropylbenzene containing tertiary allylic hydrogen atom; and 1-4-pentadiene containing super allylic secondary hydrogen atoms. The M06-2X/6-311++G(d,p) level of theory was used to optimize the geometries of all of the reactants, transition states, products and also the hinder rotation treatments for lower frequency modes. The G4 level of theory was used to calculate the electronic single point energies for those species to determine the 0 K barriers to reaction. Conventional transition state theory with Eckart tunnelling corrections was used to calculate the rate constants. The comparison between our calculated rate constants with the available experimental results from the literature shows good agreement for the reactions of propene and isobutene with molecular oxygen. The rate constant for toluene with O 2 is about an order magnitude slower than that experimentally derived from a comprehensive model proposed by Oehlschlaeger and coauthors. The results clearly indicate the need for a more detailed investigation of the combustion kinetics of toluene oxidation and its key pyrolysis and oxidation intermediates. Despite this, our computed barriers and rate constants retain an important internal consistency. Rate constants

  17. Use of Machine Learning Algorithms to Propose a New Methodology to Conduct, Critique and Validate Urban Scale Building Energy Modeling

    NASA Astrophysics Data System (ADS)

    Pathak, Maharshi

    City administrators and real-estate developers have been setting up rather aggressive energy efficiency targets. This, in turn, has led the building science research groups across the globe to focus on urban scale building performance studies and level of abstraction associated with the simulations of the same. The increasing maturity of the stakeholders towards energy efficiency and creating comfortable working environment has led researchers to develop methodologies and tools for addressing the policy driven interventions whether it's urban level energy systems, buildings' operational optimization or retrofit guidelines. Typically, these large-scale simulations are carried out by grouping buildings based on their design similarities i.e. standardization of the buildings. Such an approach does not necessarily lead to potential working inputs which can make decision-making effective. To address this, a novel approach is proposed in the present study. The principle objective of this study is to propose, to define and evaluate the methodology to utilize machine learning algorithms in defining representative building archetypes for the Stock-level Building Energy Modeling (SBEM) which are based on operational parameter database. The study uses "Phoenix- climate" based CBECS-2012 survey microdata for analysis and validation. Using the database, parameter correlations are studied to understand the relation between input parameters and the energy performance. Contrary to precedence, the study establishes that the energy performance is better explained by the non-linear models. The non-linear behavior is explained by advanced learning algorithms. Based on these algorithms, the buildings at study are grouped into meaningful clusters. The cluster "mediod" (statistically the centroid, meaning building that can be represented as the centroid of the cluster) are established statistically to identify the level of abstraction that is acceptable for the whole building energy

  18. Unsupervised machine learning account of magnetic transitions in the Hubbard model

    NASA Astrophysics Data System (ADS)

    Ch'ng, Kelvin; Vazquez, Nick; Khatami, Ehsan

    2018-01-01

    We employ several unsupervised machine learning techniques, including autoencoders, random trees embedding, and t -distributed stochastic neighboring ensemble (t -SNE), to reduce the dimensionality of, and therefore classify, raw (auxiliary) spin configurations generated, through Monte Carlo simulations of small clusters, for the Ising and Fermi-Hubbard models at finite temperatures. Results from a convolutional autoencoder for the three-dimensional Ising model can be shown to produce the magnetization and the susceptibility as a function of temperature with a high degree of accuracy. Quantum fluctuations distort this picture and prevent us from making such connections between the output of the autoencoder and physical observables for the Hubbard model. However, we are able to define an indicator based on the output of the t -SNE algorithm that shows a near perfect agreement with the antiferromagnetic structure factor of the model in two and three spatial dimensions in the weak-coupling regime. t -SNE also predicts a transition to the canted antiferromagnetic phase for the three-dimensional model when a strong magnetic field is present. We show that these techniques cannot be expected to work away from half filling when the "sign problem" in quantum Monte Carlo simulations is present.

  19. Using a Guided Machine Learning Ensemble Model to Predict Discharge Disposition following Meningioma Resection.

    PubMed

    Muhlestein, Whitney E; Akagi, Dallin S; Kallos, Justiss A; Morone, Peter J; Weaver, Kyle D; Thompson, Reid C; Chambless, Lola B

    2018-04-01

    Objective  Machine learning (ML) algorithms are powerful tools for predicting patient outcomes. This study pilots a novel approach to algorithm selection and model creation using prediction of discharge disposition following meningioma resection as a proof of concept. Materials and Methods  A diversity of ML algorithms were trained on a single-institution database of meningioma patients to predict discharge disposition. Algorithms were ranked by predictive power and top performers were combined to create an ensemble model. The final ensemble was internally validated on never-before-seen data to demonstrate generalizability. The predictive power of the ensemble was compared with a logistic regression. Further analyses were performed to identify how important variables impact the ensemble. Results  Our ensemble model predicted disposition significantly better than a logistic regression (area under the curve of 0.78 and 0.71, respectively, p  = 0.01). Tumor size, presentation at the emergency department, body mass index, convexity location, and preoperative motor deficit most strongly influence the model, though the independent impact of individual variables is nuanced. Conclusion  Using a novel ML technique, we built a guided ML ensemble model that predicts discharge destination following meningioma resection with greater predictive power than a logistic regression, and that provides greater clinical insight than a univariate analysis. These techniques can be extended to predict many other patient outcomes of interest.

  20. Machine learnt bond order potential to model metal-organic (Co-C) heterostructures.

    PubMed

    Narayanan, Badri; Chan, Henry; Kinaci, Alper; Sen, Fatih G; Gray, Stephen K; Chan, Maria K Y; Sankaranarayanan, Subramanian K R S

    2017-11-30

    A fundamental understanding of the inter-relationships between structure, morphology, atomic scale dynamics, chemistry, and physical properties of mixed metallic-covalent systems is essential to design novel functional materials for applications in flexible nano-electronics, energy storage and catalysis. To achieve such knowledge, it is imperative to develop robust and computationally efficient atomistic models that describe atomic interactions accurately within a single framework. Here, we present a unified Tersoff-Brenner type bond order potential (BOP) for a Co-C system, trained against lattice parameters, cohesive energies, equation of state, and elastic constants of different crystalline phases of cobalt as well as orthorhombic Co 2 C derived from density functional theory (DFT) calculations. The independent BOP parameters are determined using a combination of supervised machine learning (genetic algorithms) and local minimization via the simplex method. Our newly developed BOP accurately describes the structural, thermodynamic, mechanical, and surface properties of both the elemental components as well as the carbide phases, in excellent accordance with DFT calculations and experiments. Using our machine-learnt BOP potential, we performed large-scale molecular dynamics simulations to investigate the effect of metal/carbon concentration on the structure and mechanical properties of porous architectures obtained via self-assembly of cobalt nanoparticles and fullerene molecules. Such porous structures have implications in flexible electronics, where materials with high electrical conductivity and low elastic stiffness are desired. Using unsupervised machine learning (clustering), we identify the pore structure, pore-distribution, and metallic conduction pathways in self-assembled structures at different C/Co ratios. We find that as the C/Co ratio increases, the connectivity between the Co nanoparticles becomes limited, likely resulting in low electrical

  1. Machine learning based cloud mask algorithm driven by radiative transfer modeling

    NASA Astrophysics Data System (ADS)

    Chen, N.; Li, W.; Tanikawa, T.; Hori, M.; Shimada, R.; Stamnes, K. H.

    2017-12-01

    Cloud detection is a critically important first step required to derive many satellite data products. Traditional threshold based cloud mask algorithms require a complicated design process and fine tuning for each sensor, and have difficulty over snow/ice covered areas. With the advance of computational power and machine learning techniques, we have developed a new algorithm based on a neural network classifier driven by extensive radiative transfer modeling. Statistical validation results obtained by using collocated CALIOP and MODIS data show that its performance is consistent over different ecosystems and significantly better than the MODIS Cloud Mask (MOD35 C6) during the winter seasons over mid-latitude snow covered areas. Simulations using a reduced number of satellite channels also show satisfactory results, indicating its flexibility to be configured for different sensors.

  2. A computational visual saliency model based on statistics and machine learning.

    PubMed

    Lin, Ru-Je; Lin, Wei-Song

    2014-08-01

    Identifying the type of stimuli that attracts human visual attention has been an appealing topic for scientists for many years. In particular, marking the salient regions in images is useful for both psychologists and many computer vision applications. In this paper, we propose a computational approach for producing saliency maps using statistics and machine learning methods. Based on four assumptions, three properties (Feature-Prior, Position-Prior, and Feature-Distribution) can be derived and combined by a simple intersection operation to obtain a saliency map. These properties are implemented by a similarity computation, support vector regression (SVR) technique, statistical analysis of training samples, and information theory using low-level features. This technique is able to learn the preferences of human visual behavior while simultaneously considering feature uniqueness. Experimental results show that our approach performs better in predicting human visual attention regions than 12 other models in two test databases. © 2014 ARVO.

  3. Modeling of Autovariator Operation as Power Components Adjuster in Adaptive Machine Drives

    NASA Astrophysics Data System (ADS)

    Balakin, P. D.; Belkov, V. N.; Shtripling, L. O.

    2018-01-01

    Full application of the available power and stationary mode preservation for the power station (engine) operation of the transport machine under the conditions of variable external loading, are topical issues. The issues solution is possible by means of mechanical drives with the autovaried rate transfer function and nonholonomic constraint of the main driving mediums. Additional to the main motion, controlled motion of the driving mediums is formed by a variable part of the transformed power flow and is implemented by the integrated control loop, functioning only on the basis of the laws of motion. The mathematical model of the mechanical autovariator operation is developed using Gibbs function, acceleration energy; the study results are presented; on their basis, the design calculations of the autovariator driving mediums and constraints, including its automatic control loop, are possible.

  4. Reliability enumeration model for the gear in a multi-functional machine

    NASA Astrophysics Data System (ADS)

    Nasution, M. K. M.; Ambarita, H.

    2018-02-01

    The angle and direction of motion play an important role in the ability of a multifunctional machine to be able to perform the task to be charged. The movement can be a rotational action that appears to perform a round, by which the rotation can be done by connecting the generator by hand through the help of a hinge formed from two rounded surfaces. The rotation of the entire arm can be carried out by the interconnection between two surfaces having a jagged ring. This link will change according to the angle of motion, and any yeast of the serration will have a share in the success of this process, therefore a robust hand measurement model is established based on canonical provisions.

  5. Multi-fidelity machine learning models for accurate bandgap predictions of solids

    DOE PAGES

    Pilania, Ghanshyam; Gubernatis, James E.; Lookman, Turab

    2016-12-28

    Here, we present a multi-fidelity co-kriging statistical learning framework that combines variable-fidelity quantum mechanical calculations of bandgaps to generate a machine-learned model that enables low-cost accurate predictions of the bandgaps at the highest fidelity level. Additionally, the adopted Gaussian process regression formulation allows us to predict the underlying uncertainties as a measure of our confidence in the predictions. In using a set of 600 elpasolite compounds as an example dataset and using semi-local and hybrid exchange correlation functionals within density functional theory as two levels of fidelities, we demonstrate the excellent learning performance of the method against actual high fidelitymore » quantum mechanical calculations of the bandgaps. The presented statistical learning method is not restricted to bandgaps or electronic structure methods and extends the utility of high throughput property predictions in a significant way.« less

  6. Multi-fidelity machine learning models for accurate bandgap predictions of solids

    SciTech Connect

    Pilania, Ghanshyam; Gubernatis, James E.; Lookman, Turab

    Here, we present a multi-fidelity co-kriging statistical learning framework that combines variable-fidelity quantum mechanical calculations of bandgaps to generate a machine-learned model that enables low-cost accurate predictions of the bandgaps at the highest fidelity level. Additionally, the adopted Gaussian process regression formulation allows us to predict the underlying uncertainties as a measure of our confidence in the predictions. In using a set of 600 elpasolite compounds as an example dataset and using semi-local and hybrid exchange correlation functionals within density functional theory as two levels of fidelities, we demonstrate the excellent learning performance of the method against actual high fidelitymore » quantum mechanical calculations of the bandgaps. The presented statistical learning method is not restricted to bandgaps or electronic structure methods and extends the utility of high throughput property predictions in a significant way.« less

  7. Data on Support Vector Machines (SVM) model to forecast photovoltaic power.

    PubMed

    Malvoni, M; De Giorgi, M G; Congedo, P M

    2016-12-01

    The data concern the photovoltaic (PV) power, forecasted by a hybrid model that considers weather variations and applies a technique to reduce the input data size, as presented in the paper entitled "Photovoltaic forecast based on hybrid pca-lssvm using dimensionality reducted data" (M. Malvoni, M.G. De Giorgi, P.M. Congedo, 2015) [1]. The quadratic Renyi entropy criteria together with the principal component analysis (PCA) are applied to the Least Squares Support Vector Machines (LS-SVM) to predict the PV power in the day-ahead time frame. The data here shared represent the proposed approach results. Hourly PV power predictions for 1,3,6,12, 24 ahead hours and for different data reduction sizes are provided in Supplementary material.

  8. Early Colorectal Cancer Detected by Machine Learning Model Using Gender, Age, and Complete Blood Count Data.

    PubMed

    Hornbrook, Mark C; Goshen, Ran; Choman, Eran; O'Keeffe-Rosetti, Maureen; Kinar, Yaron; Liles, Elizabeth G; Rust, Kristal C

    2017-10-01

    Machine learning tools identify patients with blood counts indicating greater likelihood of colorectal cancer and warranting colonoscopy referral. To validate a machine learning colorectal cancer detection model on a US community-based insured adult population. Eligible colorectal cancer cases (439 females, 461 males) with complete blood counts before diagnosis were identified from Kaiser Permanente Northwest Region's Tumor Registry. Control patients (n = 9108) were randomly selected from KPNW's population who had no cancers, received at ≥1 blood count, had continuous enrollment from 180 days prior to the blood count through 24 months after the count, and were aged 40-89. For each control, one blood count was randomly selected as the pseudo-colorectal cancer diagnosis date for matching to cases, and assigned a "calendar year" based on the count date. For each calendar year, 18 controls were randomly selected to match the general enrollment's 10-year age groups and lengths of continuous enrollment. Prediction performance was evaluated by area under the curve, specificity, and odds ratios. Area under the receiver operating characteristics curve for detecting colorectal cancer was 0.80 ± 0.01. At 99% specificity, the odds ratio for association of a high-risk detection score with colorectal cancer was 34.7 (95% CI 28.9-40.4). The detection model had the highest accuracy in identifying right-sided colorectal cancers. ColonFlag ® identifies individuals with tenfold higher risk of undiagnosed colorectal cancer at curable stages (0/I/II), flags colorectal tumors 180-360 days prior to usual clinical diagnosis, and is more accurate at identifying right-sided (compared to left-sided) colorectal cancers.

  9. Development of a model of machine hand eye coordination and program specifications for a topological machine vision system

    NASA Technical Reports Server (NTRS)

    1972-01-01

    A unified approach to computer vision and manipulation is developed which is called choreographic vision. In the model, objects to be viewed by a projected robot in the Viking missions to Mars are seen as objects to be manipulated within choreographic contexts controlled by a multimoded remote, supervisory control system on Earth. A new theory of context relations is introduced as a basis for choreographic programming languages. A topological vision model is developed for recognizing objects by shape and contour. This model is integrated with a projected vision system consisting of a multiaperture image dissector TV camera and a ranging laser system. System program specifications integrate eye-hand coordination and topological vision functions and an aerospace multiprocessor implementation is described.

  10. Machine learning modeling of plant phenology based on coupling satellite and gridded meteorological dataset

    NASA Astrophysics Data System (ADS)

    Czernecki, Bartosz; Nowosad, Jakub; Jabłońska, Katarzyna

    2018-04-01

    Changes in the timing of plant phenological phases are important proxies in contemporary climate research. However, most of the commonly used traditional phenological observations do not give any coherent spatial information. While consistent spatial data can be obtained from airborne sensors and preprocessed gridded meteorological data, not many studies robustly benefit from these data sources. Therefore, the main aim of this study is to create and evaluate different statistical models for reconstructing, predicting, and improving quality of phenological phases monitoring with the use of satellite and meteorological products. A quality-controlled dataset of the 13 BBCH plant phenophases in Poland was collected for the period 2007-2014. For each phenophase, statistical models were built using the most commonly applied regression-based machine learning techniques, such as multiple linear regression, lasso, principal component regression, generalized boosted models, and random forest. The quality of the models was estimated using a k-fold cross-validation. The obtained results showed varying potential for coupling meteorological derived indices with remote sensing products in terms of phenological modeling; however, application of both data sources improves models' accuracy from 0.6 to 4.6 day in terms of obtained RMSE. It is shown that a robust prediction of early phenological phases is mostly related to meteorological indices, whereas for autumn phenophases, there is a stronger information signal provided by satellite-derived vegetation metrics. Choosing a specific set of predictors and applying a robust preprocessing procedures is more important for final results than the selection of a particular statistical model. The average RMSE for the best models of all phenophases is 6.3, while the individual RMSE vary seasonally from 3.5 to 10 days. Models give reliable proxy for ground observations with RMSE below 5 days for early spring and late spring phenophases. For

  11. Prediction of CO concentrations based on a hybrid Partial Least Square and Support Vector Machine model

    NASA Astrophysics Data System (ADS)

    Yeganeh, B.; Motlagh, M. Shafie Pour; Rashidi, Y.; Kamalan, H.

    2012-08-01

    Due to the health impacts caused by exposures to air pollutants in urban areas, monitoring and forecasting of air quality parameters have become popular as an important topic in atmospheric and environmental research today. The knowledge on the dynamics and complexity of air pollutants behavior has made artificial intelligence models as a useful tool for a more accurate pollutant concentration prediction. This paper focuses on an innovative method of daily air pollution prediction using combination of Support Vector Machine (SVM) as predictor and Partial Least Square (PLS) as a data selection tool based on the measured values of CO concentrations. The CO concentrations of Rey monitoring station in the south of Tehran, from Jan. 2007 to Feb. 2011, have been used to test the effectiveness of this method. The hourly CO concentrations have been predicted using the SVM and the hybrid PLS-SVM models. Similarly, daily CO concentrations have been predicted based on the aforementioned four years measured data. Results demonstrated that both models have good prediction ability; however the hybrid PLS-SVM has better accuracy. In the analysis presented in this paper, statistic estimators including relative mean errors, root mean squared errors and the mean absolute relative error have been employed to compare performances of the models. It has been concluded that the errors decrease after size reduction and coefficients of determination increase from 56 to 81% for SVM model to 65-85% for hybrid PLS-SVM model respectively. Also it was found that the hybrid PLS-SVM model required lower computational time than SVM model as expected, hence supporting the more accurate and faster prediction ability of hybrid PLS-SVM model.

  12. Modeling the control of the central nervous system over the cardiovascular system using support vector machines.

    PubMed

    Díaz, José; Acosta, Jesús; González, Rafael; Cota, Juan; Sifuentes, Ernesto; Nebot, Àngela

    2018-02-01

    The control of the central nervous system (CNS) over the cardiovascular system (CS) has been modeled using different techniques, such as fuzzy inductive reasoning, genetic fuzzy systems, neural networks, and nonlinear autoregressive techniques; the results obtained so far have been significant, but not solid enough to describe the control response of the CNS over the CS. In this research, support vector machines (SVMs) are used to predict the response of a branch of the CNS, specifically, the one that controls an important part of the cardiovascular system. To do this, five models are developed to emulate the output response of five controllers for the same input signal, the carotid sinus blood pressure (CSBP). These controllers regulate parameters such as heart rate, myocardial contractility, peripheral and coronary resistance, and venous tone. The models are trained using a known set of input-output response in each controller; also, there is a set of six input-output signals for testing each proposed model. The input signals are processed using an all-pass filter, and the accuracy performance of the control models is evaluated using the percentage value of the normalized mean square error (MSE). Experimental results reveal that SVM models achieve a better estimation of the dynamical behavior of the CNS control compared to others modeling systems. The main results obtained show that the best case is for the peripheral resistance controller, with a MSE of 1.20e-4%, while the worst case is for the heart rate controller, with a MSE of 1.80e-3%. These novel models show a great reliability in fitting the output response of the CNS which can be used as an input to the hemodynamic system models in order to predict the behavior of the heart and blood vessels in response to blood pressure variations. Copyright © 2017 Elsevier Ltd. All rights reserved.

  13. Prediction of Aerosol Optical Depth in West Asia: Machine Learning Methods versus Numerical Models

    NASA Astrophysics Data System (ADS)

    Omid Nabavi, Seyed; Haimberger, Leopold; Abbasi, Reyhaneh; Samimi, Cyrus

    2017-04-01

    Dust-prone areas of West Asia are releasing increasingly large amounts of dust particles during warm months. Because of the lack of ground-based observations in the region, this phenomenon is mainly monitored through remotely sensed aerosol products. The recent development of mesoscale Numerical Models (NMs) has offered an unprecedented opportunity to predict dust emission, and, subsequently Aerosol Optical Depth (AOD), at finer spatial and temporal resolutions. Nevertheless, the significant uncertainties in input data and simulations of dust activation and transport limit the performance of numerical models in dust prediction. The presented study aims to evaluate if machine-learning algorithms (MLAs), which require much less computational expense, can yield the same or even better performance than NMs. Deep blue (DB) AOD, which is observed by satellites but also predicted by MLAs and NMs, is used for validation. We concentrate our evaluations on the over dry Iraq plains, known as the main origin of recently intensified dust storms in West Asia. Here we examine the performance of four MLAs including Linear regression Model (LM), Support Vector Machine (SVM), Artificial Neural Network (ANN), Multivariate Adaptive Regression Splines (MARS). The Weather Research and Forecasting model coupled to Chemistry (WRF-Chem) and the Dust REgional Atmosphere Model (DREAM) are included as NMs. The MACC aerosol re-analysis of European Centre for Medium-range Weather Forecast (ECMWF) is also included, although it has assimilated satellite-based AOD data. Using the Recursive Feature Elimination (RFE) method, nine environmental features including soil moisture and temperature, NDVI, dust source function, albedo, dust uplift potential, vertical velocity, precipitation and 9-month SPEI drought index are selected for dust (AOD) modeling by MLAs. During the feature selection process, we noticed that NDVI and SPEI are of the highest importance in MLAs predictions. The data set was divided

  14. Using machine learning tools to model complex toxic interactions with limited sampling regimes.

    PubMed

    Bertin, Matthew J; Moeller, Peter; Guillette, Louis J; Chapman, Robert W

    2013-03-19

    A major impediment to understanding the impact of environmental stress, including toxins and other pollutants, on organisms, is that organisms are rarely challenged by one or a few stressors in natural systems. Thus, linking laboratory experiments that are limited by practical considerations to a few stressors and a few levels of these stressors to real world conditions is constrained. In addition, while the existence of complex interactions among stressors can be identified by current statistical methods, these methods do not provide a means to construct mathematical models of these interactions. In this paper, we offer a two-step process by which complex interactions of stressors on biological systems can be modeled in an experimental design that is within the limits of practicality. We begin with the notion that environment conditions circumscribe an n-dimensional hyperspace within which biological processes or end points are embedded. We then randomly sample this hyperspace to establish experimental conditions that span the range of the relevant parameters and conduct the experiment(s) based upon these selected conditions. Models of the complex interactions of the parameters are then extracted using machine learning tools, specifically artificial neural networks. This approach can rapidly generate highly accurate models of biological responses to complex interactions among environmentally relevant toxins, identify critical subspaces where nonlinear responses exist, and provide an expedient means of designing traditional experiments to test the impact of complex mixtures on biological responses. Further, this can be accomplished with an astonishingly small sample size.

  15. Model of Peatland Vegetation Species using HyMap Image and Machine Learning

    NASA Astrophysics Data System (ADS)

    Dayuf Jusuf, Muhammad; Danoedoro, Projo; Muljo Sukojo, Bangun; Hartono

    2017-12-01

    Species Tumih / Parepat (Combretocarpus-rotundatus Mig. Dancer) family Anisophylleaceae and Meranti (Shorea Belangerang, Shorea Teysmanniana Dyer ex Brandis) family Dipterocarpaceae is a group of vegetation species distribution model. Species pioneer is predicted as an indicator of the succession of ecosystem restoration of tropical peatland characteristics and extremely fragile (unique) in the endemic hot spot of Sundaland. Climate change projections and conservation planning are hot topics of current discussion, analysis of alternative approaches and the development of combinations of species projection modelling algorithms through geospatial information systems technology. Approach model to find out the research problem of vegetation level based on the machine learning hybrid method, wavelet and artificial neural networks. Field data are used as a reference collection of natural resource field sample objects and biodiversity assessment. The testing and training ANN data set iterations times 28, achieve a performance value of 0.0867 MSE value is smaller than the ANN training data, above 50%, and spectral accuracy 82.1 %. Identify the location of the sample point position of the Tumih / Parepat vegetation species using HyMap Image is good enough, at least the modelling, design of the species distribution can reach the target in this study. The computation validation rate above 90% proves the calculation can be considered.

  16. A Hybrid dasymetric and machine learning approach to high-resolution residential electricity consumption modeling

    SciTech Connect

    Morton, April M; Nagle, Nicholas N; Piburn, Jesse O

    As urban areas continue to grow and evolve in a world of increasing environmental awareness, the need for detailed information regarding residential energy consumption patterns has become increasingly important. Though current modeling efforts mark significant progress in the effort to better understand the spatial distribution of energy consumption, the majority of techniques are highly dependent on region-specific data sources and often require building- or dwelling-level details that are not publicly available for many regions in the United States. Furthermore, many existing methods do not account for errors in input data sources and may not accurately reflect inherent uncertainties in modelmore » outputs. We propose an alternative and more general hybrid approach to high-resolution residential electricity consumption modeling by merging a dasymetric model with a complementary machine learning algorithm. The method s flexible data requirement and statistical framework ensure that the model both is applicable to a wide range of regions and considers errors in input data sources.« less

  17. Accurate modeling of switched reluctance machine based on hybrid trained WNN

    SciTech Connect

    Song, Shoujun, E-mail: sunnyway@nwpu.edu.cn; Ge, Lefei; Ma, Shaojie

    2014-04-15

    According to the strong nonlinear electromagnetic characteristics of switched reluctance machine (SRM), a novel accurate modeling method is proposed based on hybrid trained wavelet neural network (WNN) which combines improved genetic algorithm (GA) with gradient descent (GD) method to train the network. In the novel method, WNN is trained by GD method based on the initial weights obtained per improved GA optimization, and the global parallel searching capability of stochastic algorithm and local convergence speed of deterministic algorithm are combined to enhance the training accuracy, stability and speed. Based on the measured electromagnetic characteristics of a 3-phase 12/8-pole SRM, themore » nonlinear simulation model is built by hybrid trained WNN in Matlab. The phase current and mechanical characteristics from simulation under different working conditions meet well with those from experiments, which indicates the accuracy of the model for dynamic and static performance evaluation of SRM and verifies the effectiveness of the proposed modeling method.« less

  18. Predictive modeling of dynamic fracture growth in brittle materials with machine learning

    SciTech Connect

    Moore, Bryan A.; Rougier, Esteban; O’Malley, Daniel

    We use simulation data from a high delity Finite-Discrete Element Model to build an e cient Machine Learning (ML) approach to predict fracture growth and coalescence. Our goal is for the ML approach to be used as an emulator in place of the computationally intensive high delity models in an uncertainty quanti cation framework where thousands of forward runs are required. The failure of materials with various fracture con gurations (size, orientation and the number of initial cracks) are explored and used as data to train our ML model. This novel approach has shown promise in predicting spatial (path tomore » failure) and temporal (time to failure) aspects of brittle material failure. Predictions of where dominant fracture paths formed within a material were ~85% accurate and the time of material failure deviated from the actual failure time by an average of ~16%. Additionally, the ML model achieves a reduction in computational cost by multiple orders of magnitude.« less

  19. Predictive modeling of dynamic fracture growth in brittle materials with machine learning

    DOE PAGES

    Moore, Bryan A.; Rougier, Esteban; O’Malley, Daniel; ...

    2018-02-22

    We use simulation data from a high delity Finite-Discrete Element Model to build an e cient Machine Learning (ML) approach to predict fracture growth and coalescence. Our goal is for the ML approach to be used as an emulator in place of the computationally intensive high delity models in an uncertainty quanti cation framework where thousands of forward runs are required. The failure of materials with various fracture con gurations (size, orientation and the number of initial cracks) are explored and used as data to train our ML model. This novel approach has shown promise in predicting spatial (path tomore » failure) and temporal (time to failure) aspects of brittle material failure. Predictions of where dominant fracture paths formed within a material were ~85% accurate and the time of material failure deviated from the actual failure time by an average of ~16%. Additionally, the ML model achieves a reduction in computational cost by multiple orders of magnitude.« less

  20. Evaluation of different machine learning models for predicting and mapping the susceptibility of gully erosion

    NASA Astrophysics Data System (ADS)

    Rahmati, Omid; Tahmasebipour, Nasser; Haghizadeh, Ali; Pourghasemi, Hamid Reza; Feizizadeh, Bakhtiar

    2017-12-01

    Gully erosion constitutes a serious problem for land degradation in a wide range of environments. The main objective of this research was to compare the performance of seven state-of-the-art machine learning models (SVM with four kernel types, BP-ANN, RF, and BRT) to model the occurrence of gully erosion in the Kashkan-Poldokhtar Watershed, Iran. In the first step, a gully inventory map consisting of 65 gully polygons was prepared through field surveys. Three different sample data sets (S1, S2, and S3), including both positive and negative cells (70% for training and 30% for validation), were randomly prepared to evaluate the robustness of the models. To model the gully erosion susceptibility, 12 geo-environmental factors were selected as predictors. Finally, the goodness-of-fit and prediction skill of the models were evaluated by different criteria, including efficiency percent, kappa coefficient, and the area under the ROC curves (AUC). In terms of accuracy, the RF, RBF-SVM, BRT, and P-SVM models performed excellently both in the degree of fitting and in predictive performance (AUC values well above 0.9), which resulted in accurate predictions. Therefore, these models can be used in other gully erosion studies, as they are capable of rapidly producing accurate and robust gully erosion susceptibility maps (GESMs) for decision-making and soil and water management practices. Furthermore, it was found that performance of RF and RBF-SVM for modelling gully erosion occurrence is quite stable when the learning and validation samples are changed.

  1. Toward a Progress Indicator for Machine Learning Model Building and Data Mining Algorithm Execution: A Position Paper.

    PubMed

    Luo, Gang

    2017-12-01

    For user-friendliness, many software systems offer progress indicators for long-duration tasks. A typical progress indicator continuously estimates the remaining task execution time as well as the portion of the task that has been finished. Building a machine learning model often takes a long time, but no existing machine learning software supplies a non-trivial progress indicator. Similarly, running a data mining algorithm often takes a long time, but no existing data mining software provides a nontrivial progress indicator. In this article, we consider the problem of offering progress indicators for machine learning model building and data mining algorithm execution. We discuss the goals and challenges intrinsic to this problem. Then we describe an initial framework for implementing such progress indicators and two advanced, potential uses of them, with the goal of inspiring future research on this topic.

  2. Toward a Progress Indicator for Machine Learning Model Building and Data Mining Algorithm Execution: A Position Paper

    PubMed Central

    Luo, Gang

    2017-01-01

    For user-friendliness, many software systems offer progress indicators for long-duration tasks. A typical progress indicator continuously estimates the remaining task execution time as well as the portion of the task that has been finished. Building a machine learning model often takes a long time, but no existing machine learning software supplies a non-trivial progress indicator. Similarly, running a data mining algorithm often takes a long time, but no existing data mining software provides a nontrivial progress indicator. In this article, we consider the problem of offering progress indicators for machine learning model building and data mining algorithm execution. We discuss the goals and challenges intrinsic to this problem. Then we describe an initial framework for implementing such progress indicators and two advanced, potential uses of them, with the goal of inspiring future research on this topic. PMID:29177022

  3. Improving protein-protein interactions prediction accuracy using protein evolutionary information and relevance vector machine model.

    PubMed

    An, Ji-Yong; Meng, Fan-Rong; You, Zhu-Hong; Chen, Xing; Yan, Gui-Ying; Hu, Ji-Pu

    2016-10-01

    Predicting protein-protein interactions (PPIs) is a challenging task and essential to construct the protein interaction networks, which is important for facilitating our understanding of the mechanisms of biological systems. Although a number of high-throughput technologies have been proposed to predict PPIs, there are unavoidable shortcomings, including high cost, time intensity, and inherently high false positive rates. For these reasons, many computational methods have been proposed for predicting PPIs. However, the problem is still far from being solved. In this article, we propose a novel computational method called RVM-BiGP that combines the relevance vector machine (RVM) model and Bi-gram Probabilities (BiGP) for PPIs detection from protein sequences. The major improvement includes (1) Protein sequences are represented using the Bi-gram probabilities (BiGP) feature representation on a Position Specific Scoring Matrix (PSSM), in which the protein evolutionary information is contained; (2) For reducing the influence of noise, the Principal Component Analysis (PCA) method is used to reduce the dimension of BiGP vector; (3) The powerful and robust Relevance Vector Machine (RVM) algorithm is used for classification. Five-fold cross-validation experiments executed on yeast and Helicobacter pylori datasets, which achieved very high accuracies of 94.57 and 90.57%, respectively. Experimental results are significantly better than previous methods. To further evaluate the proposed method, we compare it with the state-of-the-art support vector machine (SVM) classifier on the yeast dataset. The experimental results demonstrate that our RVM-BiGP method is significantly better than the SVM-based method. In addition, we achieved 97.15% accuracy on imbalance yeast dataset, which is higher than that of balance yeast dataset. The promising experimental results show the efficiency and robust of the proposed method, which can be an automatic decision support tool for future

  4. Development of hardware system using temperature and vibration maintenance models integration concepts for conventional machines monitoring: a case study

    NASA Astrophysics Data System (ADS)

    Adeyeri, Michael Kanisuru; Mpofu, Khumbulani; Kareem, Buliaminu

    2016-03-01

    This article describes the integration of temperature and vibration models for maintenance monitoring of conventional machinery parts in which their optimal and best functionalities are affected by abnormal changes in temperature and vibration values thereby resulting in machine failures, machines breakdown, poor quality of products, inability to meeting customers' demand, poor inventory control and just to mention a few. The work entails the use of temperature and vibration sensors as monitoring probes programmed in microcontroller using C language. The developed hardware consists of vibration sensor of ADXL345, temperature sensor of AD594/595 of type K thermocouple, microcontroller, graphic liquid crystal display, real time clock, etc. The hardware is divided into two: one is based at the workstation (majorly meant to monitor machines behaviour) and the other at the base station (meant to receive transmission of machines information sent from the workstation), working cooperatively for effective functionalities. The resulting hardware built was calibrated, tested using model verification and validated through principles pivoted on least square and regression analysis approach using data read from the gear boxes of extruding and cutting machines used for polyethylene bag production. The results got therein confirmed related correlation existing between time, vibration and temperature, which are reflections of effective formulation of the developed concept.

  5. Robust Least-Squares Support Vector Machine With Minimization of Mean and Variance of Modeling Error.

    PubMed

    Lu, Xinjiang; Liu, Wenbo; Zhou, Chuang; Huang, Minghui

    2017-06-13

    The least-squares support vector machine (LS-SVM) is a popular data-driven modeling method and has been successfully applied to a wide range of applications. However, it has some disadvantages, including being ineffective at handling non-Gaussian noise as well as being sensitive to outliers. In this paper, a robust LS-SVM method is proposed and is shown to have more reliable performance when modeling a nonlinear system under conditions where Gaussian or non-Gaussian noise is present. The construction of a new objective function allows for a reduction of the mean of the modeling error as well as the minimization of its variance, and it does not constrain the mean of the modeling error to zero. This differs from the traditional LS-SVM, which uses a worst-case scenario approach in order to minimize the modeling error and constrains the mean of the modeling error to zero. In doing so, the proposed method takes the modeling error distribution information into consideration and is thus less conservative and more robust in regards to random noise. A solving method is then developed in order to determine the optimal parameters for the proposed robust LS-SVM. An additional analysis indicates that the proposed LS-SVM gives a smaller weight to a large-error training sample and a larger weight to a small-error training sample, and is thus more robust than the traditional LS-SVM. The effectiveness of the proposed robust LS-SVM is demonstrated using both artificial and real life cases.

  6. Filtered selection coupled with support vector machines generate a functionally relevant prediction model for colorectal cancer

    PubMed Central

    Gabere, Musa Nur; Hussein, Mohamed Aly; Aziz, Mohammad Azhar

    2016-01-01

    Purpose There has been considerable interest in using whole-genome expression profiles for the classification of colorectal cancer (CRC). The selection of important features is a crucial step before training a classifier. Methods In this study, we built a model that uses support vector machine (SVM) to classify cancer and normal samples using Affymetrix exon microarray data obtained from 90 samples of 48 patients diagnosed with CRC. From the 22,011 genes, we selected the 20, 30, 50, 100, 200, 300, and 500 genes most relevant to CRC using the minimum-redundancy–maximum-relevance (mRMR) technique. With these gene sets, an SVM model was designed using four different kernel types (linear, polynomial, radial basis function [RBF], and sigmoid). Results The best model, which used 30 genes and RBF kernel, outperformed other combinations; it had an accuracy of 84% for both ten fold and leave-one-out cross validations in discriminating the cancer samples from the normal samples. With this 30 genes set from mRMR, six classifiers were trained using random forest (RF), Bayes net (BN), multilayer perceptron (MLP), naïve Bayes (NB), reduced error pruning tree (REPT), and SVM. Two hybrids, mRMR + SVM and mRMR + BN, were the best models when tested on other datasets, and they achieved a prediction accuracy of 95.27% and 91.99%, respectively, compared to other mRMR hybrid models (mRMR + RF, mRMR + NB, mRMR + REPT, and mRMR + MLP). Ingenuity pathway analysis was used to analyze the functions of the 30 genes selected for this model and their potential association with CRC: CDH3, CEACAM7, CLDN1, IL8, IL6R, MMP1, MMP7, and TGFB1 were predicted to be CRC biomarkers. Conclusion This model could be used to further develop a diagnostic tool for predicting CRC based on gene expression data from patient samples. PMID:27330311

  7. Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations

    NASA Astrophysics Data System (ADS)

    Ward, Logan; Liu, Ruoqian; Krishna, Amar; Hegde, Vinay I.; Agrawal, Ankit; Choudhary, Alok; Wolverton, Chris

    2017-07-01

    While high-throughput density functional theory (DFT) has become a prevalent tool for materials discovery, it is limited by the relatively large computational cost. In this paper, we explore using DFT data from high-throughput calculations to create faster, surrogate models with machine learning (ML) that can be used to guide new searches. Our method works by using decision tree models to map DFT-calculated formation enthalpies to a set of attributes consisting of two distinct types: (i) composition-dependent attributes of elemental properties (as have been used in previous ML models of DFT formation energies), combined with (ii) attributes derived from the Voronoi tessellation of the compound's crystal structure. The ML models created using this method have half the cross-validation error and similar training and evaluation speeds to models created with the Coulomb matrix and partial radial distribution function methods. For a dataset of 435 000 formation energies taken from the Open Quantum Materials Database (OQMD), our model achieves a mean absolute error of 80 meV/atom in cross validation, which is lower than the approximate error between DFT-computed and experimentally measured formation enthalpies and below 15% of the mean absolute deviation of the training set. We also demonstrate that our method can accurately estimate the formation energy of materials outside of the training set and be used to identify materials with especially large formation enthalpies. We propose that our models can be used to accelerate the discovery of new materials by identifying the most promising materials to study with DFT at little additional computational cost.

  8. A survey of supervised machine learning models for mobile-phone based pathogen identification and classification

    NASA Astrophysics Data System (ADS)

    Ceylan Koydemir, Hatice; Feng, Steve; Liang, Kyle; Nadkarni, Rohan; Tseng, Derek; Benien, Parul; Ozcan, Aydogan

    2017-03-01

    Giardia lamblia causes a disease known as giardiasis, which results in diarrhea, abdominal cramps, and bloating. Although conventional pathogen detection methods used in water analysis laboratories offer high sensitivity and specificity, they are time consuming, and need experts to operate bulky equipment and analyze the samples. Here we present a field-portable and cost-effective smartphone-based waterborne pathogen detection platform that can automatically classify Giardia cysts using machine learning. Our platform enables the detection and quantification of Giardia cysts in one hour, including sample collection, labeling, filtration, and automated counting steps. We evaluated the performance of three prototypes using Giardia-spiked water samples from different sources (e.g., reagent-grade, tap, non-potable, and pond water samples). We populated a training database with >30,000 cysts and estimated our detection sensitivity and specificity using 20 different classifier models, including decision trees, nearest neighbor classifiers, support vector machines (SVMs), and ensemble classifiers, and compared their speed of training and classification, as well as predicted accuracies. Among them, cubic SVM, medium Gaussian SVM, and bagged-trees were the most promising classifier types with accuracies of 94.1%, 94.2%, and 95%, respectively; we selected the latter as our preferred classifier for the detection and enumeration of Giardia cysts that are imaged using our mobile-phone fluorescence microscope. Without the need for any experts or microbiologists, this field-portable pathogen detection platform can present a useful tool for water quality monitoring in resource-limited-settings.

  9. Modeling workflow to design machine translation applications for public health practice

    PubMed Central

    Turner, Anne M.; Brownstein, Megumu K.; Cole, Kate; Karasz, Hilary; Kirchhoff, Katrin

    2014-01-01

    Objective Provide a detailed understanding of the information workflow processes related to translating health promotion materials for limited English proficiency individuals in order to inform the design of context-driven machine translation (MT) tools for public health (PH). Materials and Methods We applied a cognitive work analysis framework to investigate the translation information workflow processes of two large health departments in Washington State. Researchers conducted interviews, performed a task analysis, and validated results with PH professionals to model translation workflow and identify functional requirements for a translation system for PH. Results The study resulted in a detailed description of work related to translation of PH materials, an information workflow diagram, and a description of attitudes towards MT technology. We identified a number of themes that hold design implications for incorporating MT in PH translation practice. A PH translation tool prototype was designed based on these findings. Discussion This study underscores the importance of understanding the work context and information workflow for which systems will be designed. Based on themes and translation information workflow processes, we identified key design guidelines for incorporating MT into PH translation work. Primary amongst these is that MT should be followed by human review for translations to be of high quality and for the technology to be adopted into practice. Counclusion The time and costs of creating multilingual health promotion materials are barriers to translation. PH personnel were interested in MT's potential to improve access to low-cost translated PH materials, but expressed concerns about ensuring quality. We outline design considerations and a potential machine translation tool to best fit MT systems into PH practice. PMID:25445922

  10. An unsupervised machine learning model for discovering latent infectious diseases using social media data.

    PubMed

    Lim, Sunghoon; Tucker, Conrad S; Kumara, Soundar

    2017-02-01

    The authors of this work propose an unsupervised machine learning model that has the ability to identify real-world latent infectious diseases by mining social media data. In this study, a latent infectious disease is defined as a communicable disease that has not yet been formalized by national public health institutes and explicitly communicated to the general public. Most existing approaches to modeling infectious-disease-related knowledge discovery through social media networks are top-down approaches that are based on already known information, such as the names of diseases and their symptoms. In existing top-down approaches, necessary but unknown information, such as disease names and symptoms, is mostly unidentified in social media data until national public health institutes have formalized that disease. Most of the formalizing processes for latent infectious diseases are time consuming. Therefore, this study presents a bottom-up approach for latent infectious disease discovery in a given location without prior information, such as disease names and related symptoms. Social media messages with user and temporal information are extracted during the data preprocessing stage. An unsupervised sentiment analysis model is then presented. Users' expressions about symptoms, body parts, and pain locations are also identified from social media data. Then, symptom weighting vectors for each individual and time period are created, based on their sentiment and social media expressions. Finally, latent-infectious-disease-related information is retrieved from individuals' symptom weighting vectors. Twitter data from August 2012 to May 2013 are used to validate this study. Real electronic medical records for 104 individuals, who were diagnosed with influenza in the same period, are used to serve as ground truth validation. The results are promising, with the highest precision, recall, and F 1 score values of 0.773, 0.680, and 0.724, respectively. This work uses individuals

  11. Geometric dimension model of virtual astronaut body for ergonomic analysis of man-machine space system

    NASA Astrophysics Data System (ADS)

    Qianxiang, Zhou

    2012-07-01

    It is very important to clarify the geometric characteristic of human body segment and constitute analysis model for ergonomic design and the application of ergonomic virtual human. The typical anthropometric data of 1122 Chinese men aged 20-35 years were collected using three-dimensional laser scanner for human body. According to the correlation between different parameters, curve fitting were made between seven trunk parameters and ten body parameters with the SPSS 16.0 software. It can be concluded that hip circumference and shoulder breadth are the most important parameters in the models and the two parameters have high correlation with the others parameters of human body. By comparison with the conventional regressive curves, the present regression equation with the seven trunk parameters is more accurate to forecast the geometric dimensions of head, neck, height and the four limbs with high precision. Therefore, it is greatly valuable for ergonomic design and analysis of man-machine system.This result will be very useful to astronaut body model analysis and application.

  12. Prediction of recombinant protein overexpression in Escherichia coli using a machine learning based model (RPOLP).

    PubMed

    Habibi, Narjeskhatoon; Norouzi, Alireza; Mohd Hashim, Siti Z; Shamsir, Mohd Shahir; Samian, Razip

    2015-11-01

    Recombinant protein overexpression, an important biotechnological process, is ruled by complex biological rules which are mostly unknown, is in need of an intelligent algorithm so as to avoid resource-intensive lab-based trial and error experiments in order to determine the expression level of the recombinant protein. The purpose of this study is to propose a predictive model to estimate the level of recombinant protein overexpression for the first time in the literature using a machine learning approach based on the sequence, expression vector, and expression host. The expression host was confined to Escherichia coli which is the most popular bacterial host to overexpress recombinant proteins. To provide a handle to the problem, the overexpression level was categorized as low, medium and high. A set of features which were likely to affect the overexpression level was generated based on the known facts (e.g. gene length) and knowledge gathered from related literature. Then, a representative sub-set of features generated in the previous objective was determined using feature selection techniques. Finally a predictive model was developed using random forest classifier which was able to adequately classify the multi-class imbalanced small dataset constructed. The result showed that the predictive model provided a promising accuracy of 80% on average, in estimating the overexpression level of a recombinant protein. Copyright © 2015 Elsevier Ltd. All rights reserved.

  13. Light and short arc rubs in rotating machines: Experimental tests and modelling

    NASA Astrophysics Data System (ADS)

    Pennacchi, P.; Bachschmid, N.; Tanzi, E.

    2009-10-01

    Rotor-to-stator rub is a non-linear phenomenon which has been analyzed many times in rotordynamics literature, but very often these studies are devoted simply to highlight non-linearities, using very simple rotors, rather than to present reliable models. However, rotor-to-stator rub is actually one of the most common faults during the operation of rotating machinery. The frequency of its occurrence is increasing due to the trend of reducing the radial clearance between the seal and the rotor in modern turbine units, pumps and compressors in order to increase efficiency. Often the rub occurs between rotor and seals and the analysis of the phenomenon cannot set aside the consideration of the different relative stiffness. This paper presents some experimental results obtained by means of a test rig in which rub conditions of real machines are reproduced. In particular short arc rubs are considered and the shaft is stiffer than the obstacle. Then a model, suitable to be employed for real rotating machinery, is presented and the simulations obtained are compared with the experimental results. The model is able to reproduce the behaviour of the test rig.

  14. Machine listening intelligence

    NASA Astrophysics Data System (ADS)

    Cella, C. E.

    2017-05-01

    This manifesto paper will introduce machine listening intelligence, an integrated research framework for acoustic and musical signals modelling, based on signal processing, deep learning and computational musicology.

  15. Improving Simulations of Extreme Flows by Coupling a Physically-based Hydrologic Model with a Machine Learning Model

    NASA Astrophysics Data System (ADS)

    Mohammed, K.; Islam, A. S.; Khan, M. J. U.; Das, M. K.

    2017-12-01

    With the large number of hydrologic models presently available along with the global weather and geographic datasets, streamflows of almost any river in the world can be easily modeled. And if a reasonable amount of observed data from that river is available, then simulations of high accuracy can sometimes be performed after calibrating the model parameters against those observed data through inverse modeling. Although such calibrated models can succeed in simulating the general trend or mean of the observed flows very well, more often than not they fail to adequately simulate the extreme flows. This causes difficulty in tasks such as generating reliable projections of future changes in extreme flows due to climate change, which is obviously an important task due to floods and droughts being closely connected to people's lives and livelihoods. We propose an approach where the outputs of a physically-based hydrologic model are used as an input to a machine learning model to try and better simulate the extreme flows. To demonstrate this offline-coupling approach, the Soil and Water Assessment Tool (SWAT) was selected as the physically-based hydrologic model, the Artificial Neural Network (ANN) as the machine learning model and the Ganges-Brahmaputra-Meghna (GBM) river system as the study area. The GBM river system, located in South Asia, is the third largest in the world in terms of freshwater generated and forms the largest delta in the world. The flows of the GBM rivers were simulated separately in order to test the performance of this proposed approach in accurately simulating the extreme flows generated by different basins that vary in size, climate, hydrology and anthropogenic intervention on stream networks. Results show that by post-processing the simulated flows of the SWAT models with ANN models, simulations of extreme flows can be significantly improved. The mean absolute errors in simulating annual maximum/minimum daily flows were minimized from 4967

  16. A Hybrid Short-Term Traffic Flow Prediction Model Based on Singular Spectrum Analysis and Kernel Extreme Learning Machine

    PubMed Central

    Lin, Ciyun; Yang, Zhaosheng; Bing, Qichun; Zhou, Xiyang

    2016-01-01

    Short-term traffic flow prediction is one of the most important issues in the field of intelligent transport system (ITS). Because of the uncertainty and nonlinearity, short-term traffic flow prediction is a challenging task. In order to improve the accuracy of short-time traffic flow prediction, a hybrid model (SSA-KELM) is proposed based on singular spectrum analysis (SSA) and kernel extreme learning machine (KELM). SSA is used to filter out the noise of traffic flow time series. Then, the filtered traffic flow data is used to train KELM model, the optimal input form of the proposed model is determined by phase space reconstruction, and parameters of the model are optimized by gravitational search algorithm (GSA). Finally, case validation is carried out using the measured data of an expressway in Xiamen, China. And the SSA-KELM model is compared with several well-known prediction models, including support vector machine, extreme learning machine, and single KLEM model. The experimental results demonstrate that performance of the proposed model is superior to that of the comparison models. Apart from accuracy improvement, the proposed model is more robust. PMID:27551829

  17. A Hybrid Short-Term Traffic Flow Prediction Model Based on Singular Spectrum Analysis and Kernel Extreme Learning Machine.

    PubMed

    Shang, Qiang; Lin, Ciyun; Yang, Zhaosheng; Bing, Qichun; Zhou, Xiyang

    2016-01-01

    Short-term traffic flow prediction is one of the most important issues in the field of intelligent transport system (ITS). Because of the uncertainty and nonlinearity, short-term traffic flow prediction is a challenging task. In order to improve the accuracy of short-time traffic flow prediction, a hybrid model (SSA-KELM) is proposed based on singular spectrum analysis (SSA) and kernel extreme learning machine (KELM). SSA is used to filter out the noise of traffic flow time series. Then, the filtered traffic flow data is used to train KELM model, the optimal input form of the proposed model is determined by phase space reconstruction, and parameters of the model are optimized by gravitational search algorithm (GSA). Finally, case validation is carried out using the measured data of an expressway in Xiamen, China. And the SSA-KELM model is compared with several well-known prediction models, including support vector machine, extreme learning machine, and single KLEM model. The experimental results demonstrate that performance of the proposed model is superior to that of the comparison models. Apart from accuracy improvement, the proposed model is more robust.

  18. Modelling daily dissolved oxygen concentration using least square support vector machine, multivariate adaptive regression splines and M5 model tree

    NASA Astrophysics Data System (ADS)

    Heddam, Salim; Kisi, Ozgur

    2018-04-01

    In the present study, three types of artificial intelligence techniques, least square support vector machine (LSSVM), multivariate adaptive regression splines (MARS) and M5 model tree (M5T) are applied for modeling daily dissolved oxygen (DO) concentration using several water quality variables as inputs. The DO concentration and water quality variables data from three stations operated by the United States Geological Survey (USGS) were used for developing the three models. The water quality data selected consisted of daily measured of water temperature (TE, °C), pH (std. unit), specific conductance (SC, μS/cm) and discharge (DI cfs), are used as inputs to the LSSVM, MARS and M5T models. The three models were applied for each station separately and compared to each other. According to the results obtained, it was found that: (i) the DO concentration could be successfully estimated using the three models and (ii) the best model among all others differs from one station to another.

  19. Machine Shop Grinding Machines.

    ERIC Educational Resources Information Center

    Dunn, James

    This curriculum manual is one in a series of machine shop curriculum manuals intended for use in full-time secondary and postsecondary classes, as well as part-time adult classes. The curriculum can also be adapted to open-entry, open-exit programs. Its purpose is to equip students with basic knowledge and skills that will enable them to enter the…

  20. Curriculum Assessment Using Artificial Neural Network and Support Vector Machine Modeling Approaches: A Case Study. IR Applications. Volume 29

    ERIC Educational Resources Information Center

    Chen, Chau-Kuang

    2010-01-01

    Artificial Neural Network (ANN) and Support Vector Machine (SVM) approaches have been on the cutting edge of science and technology for pattern recognition and data classification. In the ANN model, classification accuracy can be achieved by using the feed-forward of inputs, back-propagation of errors, and the adjustment of connection weights. In…

  1. Support Vector Machine Model for Automatic Detection and Classification of Seismic Events

    NASA Astrophysics Data System (ADS)

    Barros, Vesna; Barros, Lucas

    2016-04-01

    The automated processing of multiple seismic signals to detect, localize and classify seismic events is a central tool in both natural hazards monitoring and nuclear treaty verification. However, false detections and missed detections caused by station noise and incorrect classification of arrivals are still an issue and the events are often unclassified or poorly classified. Thus, machine learning techniques can be used in automatic processing for classifying the huge database of seismic recordings and provide more confidence in the final output. Applied in the context of the International Monitoring System (IMS) - a global sensor network developed for the Comprehensive Nuclear-Test-Ban Treaty (CTBT) - we propose a fully automatic method for seismic event detection and classification based on a supervised pattern recognition technique called the Support Vector Machine (SVM). According to Kortström et al., 2015, the advantages of using SVM are handleability of large number of features and effectiveness in high dimensional spaces. Our objective is to detect seismic events from one IMS seismic station located in an area of high seismicity and mining activity and classify them as earthquakes or quarry blasts. It is expected to create a flexible and easily adjustable SVM method that can be applied in different regions and datasets. Taken a step further, accurate results for seismic stations could lead to a modification of the model and its parameters to make it applicable to other waveform technologies used to monitor nuclear explosions such as infrasound and hydroacoustic waveforms. As an authorized user, we have direct access to all IMS data and bulletins through a secure signatory account. A set of significant seismic waveforms containing different types of events (e.g. earthquake, quarry blasts) and noise is being analysed to train the model and learn the typical pattern of the signal from these events. Moreover, comparing the performance of the support

  2. Machine learning to construct reduced-order models and scaling laws for reactive-transport applications

    NASA Astrophysics Data System (ADS)

    Mudunuru, M. K.; Karra, S.; Vesselinov, V. V.

    2017-12-01

    The efficiency of many hydrogeological applications such as reactive-transport and contaminant remediation vastly depends on the macroscopic mixing occurring in the aquifer. In the case of remediation activities, it is fundamental to enhancement and control of the mixing through impact of the structure of flow field which is impacted by groundwater pumping/extraction, heterogeneity, and anisotropy of the flow medium. However, the relative importance of these hydrogeological parameters to understand mixing process is not well studied. This is partially because to understand and quantify mixing, one needs to perform multiple runs of high-fidelity numerical simulations for various subsurface model inputs. Typically, high-fidelity simulations of existing subsurface models take hours to complete on several thousands of processors. As a result, they may not be feasible to study the importance and impact of model inputs on mixing. Hence, there is a pressing need to develop computationally efficient models to accurately predict the desired QoIs for remediation and reactive-transport applications. An attractive way to construct computationally efficient models is through reduced-order modeling using machine learning. These approaches can substantially improve our capabilities to model and predict remediation process. Reduced-Order Models (ROMs) are similar to analytical solutions or lookup tables. However, the method in which ROMs are constructed is different. Here, we present a physics-informed ML framework to construct ROMs based on high-fidelity numerical simulations. First, random forests, F-test, and mutual information are used to evaluate the importance of model inputs. Second, SVMs are used to construct ROMs based on these inputs. These ROMs are then used to understand mixing under perturbed vortex flows. Finally, we construct scaling laws for certain important QoIs such as degree of mixing and product yield. Scaling law parameters dependence on model inputs are

  3. Coupling machine learning with mechanistic models to study runoff production and river flow at the hillslope scale

    NASA Astrophysics Data System (ADS)

    Marçais, J.; Gupta, H. V.; De Dreuzy, J. R.; Troch, P. A. A.

    2016-12-01

    Geomorphological structure and geological heterogeneity of hillslopes are major controls on runoff responses. The diversity of hillslopes (morphological shapes and geological structures) on one hand, and the highly non linear runoff mechanism response on the other hand, make it difficult to transpose what has been learnt at one specific hillslope to another. Therefore, making reliable predictions on runoff appearance or river flow for a given hillslope is a challenge. Applying a classic model calibration (based on inverse problems technique) requires doing it for each specific hillslope and having some data available for calibration. When applied to thousands of cases it cannot always be promoted. Here we propose a novel modeling framework based on coupling process based models with data based approach. First we develop a mechanistic model, based on hillslope storage Boussinesq equations (Troch et al. 2003), able to model non linear runoff responses to rainfall at the hillslope scale. Second we set up a model database, representing thousands of non calibrated simulations. These simulations investigate different hillslope shapes (real ones obtained by analyzing 5m digital elevation model of Brittany and synthetic ones), different hillslope geological structures (i.e. different parametrizations) and different hydrologic forcing terms (i.e. different infiltration chronicles). Then, we use this model library to train a machine learning model on this physically based database. Machine learning model performance is then assessed by a classic validating phase (testing it on new hillslopes and comparing machine learning with mechanistic outputs). Finally we use this machine learning model to learn what are the hillslope properties controlling runoffs. This methodology will be further tested combining synthetic datasets with real ones.

  4. Assessing biomass of diverse coastal marsh ecosystems using statistical and machine learning models

    NASA Astrophysics Data System (ADS)

    Mo, Yu; Kearney, Michael S.; Riter, J. C. Alexis; Zhao, Feng; Tilley, David R.

    2018-06-01

    The importance and vulnerability of coastal marshes necessitate effective ways to closely monitor them. Optical remote sensing is a powerful tool for this task, yet its application to diverse coastal marsh ecosystems consisting of different marsh types is limited. This study samples spectral and biophysical data from freshwater, intermediate, brackish, and saline marshes in Louisiana, and develops statistical and machine learning models to assess the marshes' biomass with combined ground, airborne, and spaceborne remote sensing data. It is found that linear models derived from NDVI and EVI are most favorable for assessing Leaf Area Index (LAI) using multispectral data (R2 = 0.7 and 0.67, respectively), and the random forest models are most useful in retrieving LAI and Aboveground Green Biomass (AGB) using hyperspectral data (R2 = 0.91 and 0.84, respectively). It is also found that marsh type and plant species significantly impact the linear model development (P < .05 in both cases). Sensors with coarser spatial resolution yield lower LAI values because the fine water networks are not detected and mixed into the vegetation pixels. The Landsat OLI-derived map shows the LAI of coastal mashes in Louisiana mostly ranges from 0 to 5.0, and is highest for freshwater marshes and for marshes in the Atchafalaya Bay delta. The CASI-derived maps show that LAI of saline marshes at Bay Batiste typically ranges from 0.9 to 1.5, and the AGB is mostly less than 900 g/m2. This study provides solutions for assessing the biomass of Louisiana's coastal marshes using various optical remote sensing techniques, and highlights the impacts of the marshes' species composition on the model development and the sensors' spatial resolution on biomass mapping, thereby providing useful tools for monitoring the biomass of coastal marshes in Louisiana and diverse coastal marsh ecosystems elsewhere.

  5. Quantitative modeling of peptide binding to TAP using support vector machine.

    PubMed

    Diez-Rivero, Carmen M; Chenlo, Bernardo; Zuluaga, Pilar; Reche, Pedro A

    2010-01-01

    The transport of peptides to the endoplasmic reticulum by the transporter associated with antigen processing (TAP) is a necessary step towards determining CD8 T cell epitopes. In this work, we have studied the predictive performance of support vector machine models trained on single residue positions and residue combinations drawn from a large dataset consisting of 613 nonamer peptides of known affinity to TAP. Predictive performance of these TAP affinity models was evaluated under 10-fold cross-validation experiments and measured using Pearson's correlation coefficients (R(p)). Our results show that every peptide position (P1-P9) contributes to TAP binding (minimum R(p) of 0.26 +/- 0.11 was achieved by a model trained on the P6 residue), although the largest contributions to binding correspond to the C-terminal end (R(p) = 0.68 +/- 0.06) and the P1 (R(p) = 0.51 +/- 0.09) and P2 (0.57 +/- 0.08) residues of the peptide. Training the models on additional peptide residues generally improved their predictive performance and a maximum correlation (R(p) = 0.89 +/- 0.03) was achieved by a model trained on the full-length sequences or a residue selection consisting of the first 5 N- and last 3 C-terminal residues of the peptides included in the training set. A system for predicting the binding affinity of peptides to TAP using the methods described here is readily available for free public use at http://imed.med.ucm.es/Tools/tapreg/. (c) 2009 Wiley-Liss, Inc.

  6. Estimating the domain of applicability for machine learning QSAR models: a study on aqueous solubility of drug discovery molecules

    NASA Astrophysics Data System (ADS)

    Schroeter, Timon Sebastian; Schwaighofer, Anton; Mika, Sebastian; Ter Laak, Antonius; Suelzle, Detlev; Ganzer, Ursula; Heinrich, Nikolaus; Müller, Klaus-Robert

    2007-12-01

    We investigate the use of different Machine Learning methods to construct models for aqueous solubility. Models are based on about 4000 compounds, including an in-house set of 632 drug discovery molecules of Bayer Schering Pharma. For each method, we also consider an appropriate method to obtain error bars, in order to estimate the domain of applicability (DOA) for each model. Here, we investigate error bars from a Bayesian model (Gaussian Process (GP)), an ensemble based approach (Random Forest), and approaches based on the Mahalanobis distance to training data (for Support Vector Machine and Ridge Regression models). We evaluate all approaches in terms of their prediction accuracy (in cross-validation, and on an external validation set of 536 molecules) and in how far the individual error bars can faithfully represent the actual prediction error.

  7. Estimating the domain of applicability for machine learning QSAR models: a study on aqueous solubility of drug discovery molecules

    NASA Astrophysics Data System (ADS)

    Schroeter, Timon Sebastian; Schwaighofer, Anton; Mika, Sebastian; Ter Laak, Antonius; Suelzle, Detlev; Ganzer, Ursula; Heinrich, Nikolaus; Müller, Klaus-Robert

    2007-09-01

    We investigate the use of different Machine Learning methods to construct models for aqueous solubility. Models are based on about 4000 compounds, including an in-house set of 632 drug discovery molecules of Bayer Schering Pharma. For each method, we also consider an appropriate method to obtain error bars, in order to estimate the domain of applicability (DOA) for each model. Here, we investigate error bars from a Bayesian model (Gaussian Process (GP)), an ensemble based approach (Random Forest), and approaches based on the Mahalanobis distance to training data (for Support Vector Machine and Ridge Regression models). We evaluate all approaches in terms of their prediction accuracy (in cross-validation, and on an external validation set of 536 molecules) and in how far the individual error bars can faithfully represent the actual prediction error.

  8. AutoQSAR: an automated machine learning tool for best-practice quantitative structure-activity relationship modeling.

    PubMed

    Dixon, Steven L; Duan, Jianxin; Smith, Ethan; Von Bargen, Christopher D; Sherman, Woody; Repasky, Matthew P

    2016-10-01

    We introduce AutoQSAR, an automated machine-learning application to build, validate and deploy quantitative structure-activity relationship (QSAR) models. The process of descriptor generation, feature selection and the creation of a large number of QSAR models has been automated into a single workflow within AutoQSAR. The models are built using a variety of machine-learning methods, and each model is scored using a novel approach. Effectiveness of the method is demonstrated through comparison with literature QSAR models using identical datasets for six end points: protein-ligand binding affinity, solubility, blood-brain barrier permeability, carcinogenicity, mutagenicity and bioaccumulation in fish. AutoQSAR demonstrates similar or better predictive performance as compared with published results for four of the six endpoints while requiring minimal human time and expertise.

  9. Estimating the domain of applicability for machine learning QSAR models: a study on aqueous solubility of drug discovery molecules.

    PubMed

    Schroeter, Timon Sebastian; Schwaighofer, Anton; Mika, Sebastian; Ter Laak, Antonius; Suelzle, Detlev; Ganzer, Ursula; Heinrich, Nikolaus; Müller, Klaus-Robert

    2007-12-01

    We investigate the use of different Machine Learning methods to construct models for aqueous solubility. Models are based on about 4000 compounds, including an in-house set of 632 drug discovery molecules of Bayer Schering Pharma. For each method, we also consider an appropriate method to obtain error bars, in order to estimate the domain of applicability (DOA) for each model. Here, we investigate error bars from a Bayesian model (Gaussian Process (GP)), an ensemble based approach (Random Forest), and approaches based on the Mahalanobis distance to training data (for Support Vector Machine and Ridge Regression models). We evaluate all approaches in terms of their prediction accuracy (in cross-validation, and on an external validation set of 536 molecules) and in how far the individual error bars can faithfully represent the actual prediction error.

  10. Estimating the domain of applicability for machine learning QSAR models: a study on aqueous solubility of drug discovery molecules.

    PubMed

    Schroeter, Timon Sebastian; Schwaighofer, Anton; Mika, Sebastian; Ter Laak, Antonius; Suelzle, Detlev; Ganzer, Ursula; Heinrich, Nikolaus; Müller, Klaus-Robert

    2007-09-01

    We investigate the use of different Machine Learning methods to construct models for aqueous solubility. Models are based on about 4000 compounds, including an in-house set of 632 drug discovery molecules of Bayer Schering Pharma. For each method, we also consider an appropriate method to obtain error bars, in order to estimate the domain of applicability (DOA) for each model. Here, we investigate error bars from a Bayesian model (Gaussian Process (GP)), an ensemble based approach (Random Forest), and approaches based on the Mahalanobis distance to training data (for Support Vector Machine and Ridge Regression models). We evaluate all approaches in terms of their prediction accuracy (in cross-validation, and on an external validation set of 536 molecules) and in how far the individual error bars can faithfully represent the actual prediction error.

  11. Predicting treatment outcome of drug-susceptible tuberculosis patients using machine-learning models.

    PubMed

    Hussain, Owais A; Junejo, Khurum N

    2018-02-20

    Tuberculosis (TB) is a deadly contagious disease and a serious global health problem. It is curable but due to its lengthy treatment process, a patient is likely to leave the treatment incomplete, leading to a more lethal, drug resistant form of disease. The World Health Organization (WHO) propagates Directly Observed Therapy Short-course (DOTS) as an effective way to stop the spread of TB in communities with a high burden. But DOTS also adds a significant burden on the financial feasibility of the program. We aim to facilitate TB programs by predicting the outcome of the treatment of a particular patient at the start of treatment so that their health workers can be utilized in a targeted and cost-effective way. The problem was modeled as a classification problem, and the outcome of treatment was predicted using state-of-art implementations of 3 machine learning algorithms. 4213 patients were evaluated, out of which 64.37% completed their treatment. Results were evaluated using 4 performance measures; accuracy, precision, sensitivity, and specificity. The models offer an improvement of more than 12% accuracy over the baseline prediction. Empirical results also revealed some insights to improve TB programs. Overall, our proposed methodology will may help teams running TB programs manage their human resources more effectively, thus saving more lives.

  12. The Use of an Acellular Oxygen Carrier in a Human Liver Model of Normothermic Machine Perfusion.

    PubMed

    Laing, Richard W; Bhogal, Ricky H; Wallace, Lorraine; Boteon, Yuri; Neil, Desley A H; Smith, Amanda; Stephenson, Barney T F; Schlegel, Andrea; Hübscher, Stefan G; Mirza, Darius F; Afford, Simon C; Mergental, Hynek

    2017-11-01

    Normothermic machine perfusion of the liver (NMP-L) is a novel technique that preserves liver grafts under near-physiological conditions while maintaining their normal metabolic activity. This process requires an adequate oxygen supply, typically delivered by packed red blood cells (RBC). We present the first experience using an acellular hemoglobin-based oxygen carrier (HBOC) Hemopure in a human model of NMP-L. Five discarded high-risk human livers were perfused with HBOC-based perfusion fluid and matched to 5 RBC-perfused livers. Perfusion parameters, oxygen extraction, metabolic activity, and histological features were compared during 6 hours of NMP-L. The cytotoxicity of Hemopure was also tested on human hepatic primary cell line cultures using an in vitro model of ischemia reperfusion injury. The vascular flow parameters and the perfusate lactate clearance were similar in both groups. The HBOC-perfused livers extracted more oxygen than those perfused with RBCs (O2 extraction ratio 13.75 vs 9.43 % ×10 per gram of tissue, P = 0.001). In vitro exposure to Hemopure did not alter intracellular levels of reactive oxygen species, and there was no increase in apoptosis or necrosis observed in any of the tested cell lines. Histological findings were comparable between groups. There was no evidence of histological damage caused by Hemopure. Hemopure can be used as an alternative oxygen carrier to packed red cells in NMP-L perfusion fluid.

  13. An Improved Model Predictive Current Controller of Switched Reluctance Machines Using Time-Multiplexed Current Sensor

    PubMed Central

    Li, Bingchu; Ling, Xiao; Huang, Yixiang; Gong, Liang; Liu, Chengliang

    2017-01-01

    This paper presents a fixed-switching-frequency model predictive current controller using multiplexed current sensor for switched reluctance machine (SRM) drives. The converter was modified to distinguish currents from simultaneously excited phases during the sampling period. The only current sensor installed in the converter was time division multiplexing for phase current sampling. During the commutation stage, the control steps of adjacent phases were shifted so that sampling time was staggered. The maximum and minimum duty ratio of pulse width modulation (PWM) was limited to keep enough sampling time for analog-to-digital (A/D) conversion. Current sensor multiplexing was realized without complex adjustment of either driver circuit nor control algorithms, while it helps to reduce the cost and errors introduced in current sampling due to inconsistency between sensors. The proposed controller is validated by both simulation and experimental results with a 1.5 kW three-phase 12/8 SRM. Satisfied current sampling is received with little difference compared with independent phase current sensors for each phase. The proposed controller tracks the reference current profile as accurately as the model predictive current controller with independent phase current sensors, while having minor tracking errors compared with a hysteresis current controller. PMID:28513554

  14. The Use of an Acellular Oxygen Carrier in a Human Liver Model of Normothermic Machine Perfusion

    PubMed Central

    Wallace, Lorraine; Boteon, Yuri; Neil, Desley AH; Smith, Amanda; Stephenson, Barney TF; Schlegel, Andrea; Hübscher, Stefan G; Mirza, Darius F

    2017-01-01

    Background Normothermic machine perfusion of the liver (NMP-L) is a novel technique that preserves liver grafts under near-physiological conditions whilst maintaining their normal metabolic activity. This process requires an adequate oxygen supply, typically delivered by packed red blood cells (RBC). We present the first experience using an acellular hemoglobin-based oxygen carrier (HBOC) Hemopure in a human model of NMP-L. Methods Five discarded high-risk human livers were perfused with HBOC-based perfusion fluid and matched to 5 RBC-perfused livers. Perfusion parameters, oxygen extraction, metabolic activity and histological features were compared during 6 hours of NMP-L. The cytotoxicity of Hemopure was also tested on human hepatic primary cell line cultures using an in vitro model of ischemia reperfusion injury. Results The vascular flow parameters and the perfusate lactate clearance were similar in both groups. The HBOC-perfused livers extracted more oxygen than those perfused with RBCs (O2ER 13.75 vs 9.43 % x105 per gram of tissue, p=0.001). In vitro exposure to Hemopure did not alter intracellular levels of reactive oxygen species and there was no increase in apoptosis or necrosis observed in any of the tested cell lines. Histological findings were comparable between groups. There was no evidence of histological damage caused by Hemopure. Conclusion Hemopure can be used as an alternative oxygen carrier to packed red cells in NMP-L perfusion fluid. PMID:28520579

  15. Metabolomic Perfusate Analysis during Kidney Machine Perfusion: The Pig Provides an Appropriate Model for Human Studies

    PubMed Central

    Nath, Jay; Guy, Alison; Smith, Thomas B.; Cobbold, Mark; Inston, Nicholas G.; Hodson, James; Tennant, Daniel A.

    2014-01-01

    Introduction Hypothermic machine perfusion offers great promise in kidney transplantation and experimental studies are needed to establish the optimal conditions for this to occur. Pig kidneys are considered to be a good model for this purpose and share many properties with human organs. However it is not established whether the metabolism of pig kidneys in such hypothermic hypoxic conditions is comparable to human organs. Methods Standard criteria human (n = 12) and porcine (n = 10) kidneys underwent HMP using the LifePort Kidney Transporter 1.0 (Organ Recovery Systems) using KPS-1 solution. Perfusate was sampled at 45 minutes and 4 hours of perfusion and metabolomic analysis performed using 1-D 1H-NMR spectroscopy. Results There was no inter-species difference in the number of metabolites identified. Of the 30 metabolites analysed, 16 (53.3%) were present in comparable concentrations in the pig and human kidney perfusates. The rate of change of concentration for 3-Hydroxybutyrate was greater for human kidneys (p<0.001). For the other 29 metabolites (96.7%), there was no difference in the rate of change of concentration between pig and human samples. Conclusions Whilst there are some differences between pig and human kidneys during HMP they appear to be metabolically similar and the pig seems to be a valid model for human studies. PMID:25502759

  16. Manifest: A computer program for 2-D flow modeling in Stirling machines

    NASA Technical Reports Server (NTRS)

    Gedeon, David

    1989-01-01

    A computer program named Manifest is discussed. Manifest is a program one might want to use to model the fluid dynamics in the manifolds commonly found between the heat exchangers and regenerators of Stirling machines; but not just in the manifolds - in the regenerators as well. And in all sorts of other places too, such as: in heaters or coolers, or perhaps even in cylinder spaces. There are probably nonStirling uses for Manifest also. In broad strokes, Manifest will: (1) model oscillating internal compressible laminar fluid flow in a wide range of two-dimensional regions, either filled with porous materials or empty; (2) present a graphics-based user-friendly interface, allowing easy selection and modification of region shape and boundary condition specification; (3) run on a personal computer, or optionally (in the case of its number-crunching module) on a supercomputer; and (4) allow interactive examination of the solution output so the user can view vector plots of flow velocity, contour plots of pressure and temperature at various locations and tabulate energy-related integrals of interest.

  17. Hidden Markov Model and Support Vector Machine based decoding of finger movements using Electrocorticography

    PubMed Central

    Wissel, Tobias; Pfeiffer, Tim; Frysch, Robert; Knight, Robert T.; Chang, Edward F.; Hinrichs, Hermann; Rieger, Jochem W.; Rose, Georg

    2013-01-01

    Objective Support Vector Machines (SVM) have developed into a gold standard for accurate classification in Brain-Computer-Interfaces (BCI). The choice of the most appropriate classifier for a particular application depends on several characteristics in addition to decoding accuracy. Here we investigate the implementation of Hidden Markov Models (HMM)for online BCIs and discuss strategies to improve their performance. Approach We compare the SVM, serving as a reference, and HMMs for classifying discrete finger movements obtained from the Electrocorticograms of four subjects doing a finger tapping experiment. The classifier decisions are based on a subset of low-frequency time domain and high gamma oscillation features. Main results We show that decoding optimization between the two approaches is due to the way features are extracted and selected and less dependent on the classifier. An additional gain in HMM performance of up to 6% was obtained by introducing model constraints. Comparable accuracies of up to 90% were achieved with both SVM and HMM with the high gamma cortical response providing the most important decoding information for both techniques. Significance We discuss technical HMM characteristics and adaptations in the context of the presented data as well as for general BCI applications. Our findings suggest that HMMs and their characteristics are promising for efficient online brain-computer interfaces. PMID:24045504

  18. Machine Learning

    NASA Astrophysics Data System (ADS)

    Hoffmann, Achim; Mahidadia, Ashesh

    The purpose of this chapter is to present fundamental ideas and techniques of machine learning suitable for the field of this book, i.e., for automated scientific discovery. The chapter focuses on those symbolic machine learning methods, which produce results that are suitable to be interpreted and understood by humans. This is particularly important in the context of automated scientific discovery as the scientific theories to be produced by machines are usually meant to be interpreted by humans. This chapter contains some of the most influential ideas and concepts in machine learning research to give the reader a basic insight into the field. After the introduction in Sect. 1, general ideas of how learning problems can be framed are given in Sect. 2. The section provides useful perspectives to better understand what learning algorithms actually do. Section 3 presents the Version space model which is an early learning algorithm as well as a conceptual framework, that provides important insight into the general mechanisms behind most learning algorithms. In section 4, a family of learning algorithms, the AQ family for learning classification rules is presented. The AQ family belongs to the early approaches in machine learning. The next, Sect. 5 presents the basic principles of decision tree learners. Decision tree learners belong to the most influential class of inductive learning algorithms today. Finally, a more recent group of learning systems are presented in Sect. 6, which learn relational concepts within the framework of logic programming. This is a particularly interesting group of learning systems since the framework allows also to incorporate background knowledge which may assist in generalisation. Section 7 discusses Association Rules - a technique that comes from the related field of Data mining. Section 8 presents the basic idea of the Naive Bayesian Classifier. While this is a very popular learning technique, the learning result is not well suited for

  19. Predicting knee replacement damage in a simulator machine using a computational model with a consistent wear factor.

    PubMed

    Zhao, Dong; Sakoda, Hideyuki; Sawyer, W Gregory; Banks, Scott A; Fregly, Benjamin J

    2008-02-01

    Wear of ultrahigh molecular weight polyethylene remains a primary factor limiting the longevity of total knee replacements (TKRs). However, wear testing on a simulator machine is time consuming and expensive, making it impractical for iterative design purposes. The objectives of this paper were first, to evaluate whether a computational model using a wear factor consistent with the TKR material pair can predict accurate TKR damage measured in a simulator machine, and second, to investigate how choice of surface evolution method (fixed or variable step) and material model (linear or nonlinear) affect the prediction. An iterative computational damage model was constructed for a commercial knee implant in an AMTI simulator machine. The damage model combined a dynamic contact model with a surface evolution model to predict how wear plus creep progressively alter tibial insert geometry over multiple simulations. The computational framework was validated by predicting wear in a cylinder-on-plate system for which an analytical solution was derived. The implant damage model was evaluated for 5 million cycles of simulated gait using damage measurements made on the same implant in an AMTI machine. Using a pin-on-plate wear factor for the same material pair as the implant, the model predicted tibial insert wear volume to within 2% error and damage depths and areas to within 18% and 10% error, respectively. Choice of material model had little influence, while inclusion of surface evolution affected damage depth and area but not wear volume predictions. Surface evolution method was important only during the initial cycles, where variable step was needed to capture rapid geometry changes due to the creep. Overall, our results indicate that accurate TKR damage predictions can be made with a computational model using a constant wear factor obtained from pin-on-plate tests for the same material pair, and furthermore, that surface evolution method matters only during the initial

  20. Landslide susceptibility modeling applying machine learning methods: A case study from Longju in the Three Gorges Reservoir area, China

    NASA Astrophysics Data System (ADS)

    Zhou, Chao; Yin, Kunlong; Cao, Ying; Ahmed, Bayes; Li, Yuanyao; Catani, Filippo; Pourghasemi, Hamid Reza

    2018-03-01

    Landslide is a common natural hazard and responsible for extensive damage and losses in mountainous areas. In this study, Longju in the Three Gorges Reservoir area in China was taken as a case study for landslide susceptibility assessment in order to develop effective risk prevention and mitigation strategies. To begin, 202 landslides were identified, including 95 colluvial landslides and 107 rockfalls. Twelve landslide causal factor maps were prepared initially, and the relationship between these factors and each landslide type was analyzed using the information value model. Later, the unimportant factors were selected and eliminated using the information gain ratio technique. The landslide locations were randomly divided into two groups: 70% for training and 30% for verifying. Two machine learning models: the support vector machine (SVM) and artificial neural network (ANN), and a multivariate statistical model: the logistic regression (LR), were applied for landslide susceptibility modeling (LSM) for each type. The LSM index maps, obtained from combining the assessment results of the two landslide types, were classified into five levels. The performance of the LSMs was evaluated using the receiver operating characteristics curve and Friedman test. Results show that the elimination of noise-generating factors and the separated modeling of each landslide type have significantly increased the prediction accuracy. The machine learning models outperformed the multivariate statistical model and SVM model was found ideal for the case study area.

  1. Hybrid polylingual object model: an efficient and seamless integration of Java and native components on the Dalvik virtual machine.

    PubMed

    Huang, Yukun; Chen, Rong; Wei, Jingbo; Pei, Xilong; Cao, Jing; Prakash Jayaraman, Prem; Ranjan, Rajiv

    2014-01-01

    JNI in the Android platform is often observed with low efficiency and high coding complexity. Although many researchers have investigated the JNI mechanism, few of them solve the efficiency and the complexity problems of JNI in the Android platform simultaneously. In this paper, a hybrid polylingual object (HPO) model is proposed to allow a CAR object being accessed as a Java object and as vice in the Dalvik virtual machine. It is an acceptable substitute for JNI to reuse the CAR-compliant components in Android applications in a seamless and efficient way. The metadata injection mechanism is designed to support the automatic mapping and reflection between CAR objects and Java objects. A prototype virtual machine, called HPO-Dalvik, is implemented by extending the Dalvik virtual machine to support the HPO model. Lifespan management, garbage collection, and data type transformation of HPO objects are also handled in the HPO-Dalvik virtual machine automatically. The experimental result shows that the HPO model outweighs the standard JNI in lower overhead on native side, better executing performance with no JNI bridging code being demanded.

  2. Diagnostic Machine Learning Models for Acute Abdominal Pain: Towards an e-Learning Tool for Medical Students.

    PubMed

    Khumrin, Piyapong; Ryan, Anna; Judd, Terry; Verspoor, Karin

    2017-01-01

    Computer-aided learning systems (e-learning systems) can help medical students gain more experience with diagnostic reasoning and decision making. Within this context, providing feedback that matches students' needs (i.e. personalised feedback) is both critical and challenging. In this paper, we describe the development of a machine learning model to support medical students' diagnostic decisions. Machine learning models were trained on 208 clinical cases presenting with abdominal pain, to predict five diagnoses. We assessed which of these models are likely to be most effective for use in an e-learning tool that allows students to interact with a virtual patient. The broader goal is to utilise these models to generate personalised feedback based on the specific patient information requested by students and their active diagnostic hypotheses.

  3. Fullrmc, a rigid body Reverse Monte Carlo modeling package enabled with machine learning and artificial intelligence.

    PubMed

    Aoun, Bachir

    2016-05-05

    A new Reverse Monte Carlo (RMC) package "fullrmc" for atomic or rigid body and molecular, amorphous, or crystalline materials is presented. fullrmc main purpose is to provide a fully modular, fast and flexible software, thoroughly documented, complex molecules enabled, written in a modern programming language (python, cython, C and C++ when performance is needed) and complying to modern programming practices. fullrmc approach in solving an atomic or molecular structure is different from existing RMC algorithms and software. In a nutshell, traditional RMC methods and software randomly adjust atom positions until the whole system has the greatest consistency with a set of experimental data. In contrast, fullrmc applies smart moves endorsed with reinforcement machine learning to groups of atoms. While fullrmc allows running traditional RMC modeling, the uniqueness of this approach resides in its ability to customize grouping atoms in any convenient way with no additional programming efforts and to apply smart and more physically meaningful moves to the defined groups of atoms. In addition, fullrmc provides a unique way with almost no additional computational cost to recur a group's selection, allowing the system to go out of local minimas by refining a group's position or exploring through and beyond not allowed positions and energy barriers the unrestricted three dimensional space around a group. © 2016 Wiley Periodicals, Inc.

  4. Fullrmc, a rigid body reverse monte carlo modeling package enabled with machine learning and artificial intelligence

    DOE PAGES

    Aoun, Bachir

    2016-01-22

    Here, a new Reverse Monte Carlo (RMC) package ‘fullrmc’ for atomic or rigid body and molecular, amorphous or crystalline materials is presented. fullrmc main purpose is to provide a fully modular, fast and flexible software, thoroughly documented, complex molecules enabled, written in a modern programming language (python, cython ,C and C++ when performance is needed) and complying to modern programming practices. fullrmc approach in solving an atomic or molecular structure is different from existing RMC algorithms and software. In a nutshell, traditional RMC methods and software randomly adjust atom positions until the whole system has the greatest consistency with amore » set of experimental data. In contrast, fullrmc applies smart moves endorsed with reinforcement machine learning to groups of atoms. While fullrmc allows running traditional RMC modelling, the uniqueness of this approach resides in its ability to customize grouping atoms in any convenient way with no additional programming efforts and to apply smart and more physically meaningful moves to the defined groups of atoms. Also fullrmc provides a unique way with almost no additional computational cost to recur a group’s selection, allowing the system to go out of local minimas by refining a group’s position or exploring through and beyond not allowed positions and energy barriers the unrestricted three dimensional space around a group.« less

  5. Fullrmc, a rigid body reverse monte carlo modeling package enabled with machine learning and artificial intelligence

    SciTech Connect

    Aoun, Bachir

    Here, a new Reverse Monte Carlo (RMC) package ‘fullrmc’ for atomic or rigid body and molecular, amorphous or crystalline materials is presented. fullrmc main purpose is to provide a fully modular, fast and flexible software, thoroughly documented, complex molecules enabled, written in a modern programming language (python, cython ,C and C++ when performance is needed) and complying to modern programming practices. fullrmc approach in solving an atomic or molecular structure is different from existing RMC algorithms and software. In a nutshell, traditional RMC methods and software randomly adjust atom positions until the whole system has the greatest consistency with amore » set of experimental data. In contrast, fullrmc applies smart moves endorsed with reinforcement machine learning to groups of atoms. While fullrmc allows running traditional RMC modelling, the uniqueness of this approach resides in its ability to customize grouping atoms in any convenient way with no additional programming efforts and to apply smart and more physically meaningful moves to the defined groups of atoms. Also fullrmc provides a unique way with almost no additional computational cost to recur a group’s selection, allowing the system to go out of local minimas by refining a group’s position or exploring through and beyond not allowed positions and energy barriers the unrestricted three dimensional space around a group.« less

  6. Application of machine learning algorithms for clinical predictive modeling: a data-mining approach in SCT.

    PubMed

    Shouval, R; Bondi, O; Mishan, H; Shimoni, A; Unger, R; Nagler, A

    2014-03-01

    Data collected from hematopoietic SCT (HSCT) centers are becoming more abundant and complex owing to the formation of organized registries and incorporation of biological data. Typically, conventional statistical methods are used for the development of outcome prediction models and risk scores. However, these analyses carry inherent properties limiting their ability to cope with large data sets with multiple variables and samples. Machine learning (ML), a field stemming from artificial intelligence, is part of a wider approach for data analysis termed data mining (DM). It enables prediction in complex data scenarios, familiar to practitioners and researchers. Technological and commercial applications are all around us, gradually entering clinical research. In the following review, we would like to expose hematologists and stem cell transplanters to the concepts, clinical applications, strengths and limitations of such methods and discuss current research in HSCT. The aim of this review is to encourage utilization of the ML and DM techniques in the field of HSCT, including prediction of transplantation outcome and donor selection.

  7. Implications of the Turing machine model of computation for processor and programming language design

    NASA Astrophysics Data System (ADS)

    Hunter, Geoffrey

    2004-01-01

    A computational process is classified according to the theoretical model that is capable of executing it; computational processes that require a non-predeterminable amount of intermediate storage for their execution are Turing-machine (TM) processes, while those whose storage are predeterminable are Finite Automation (FA) processes. Simple processes (such as traffic light controller) are executable by Finite Automation, whereas the most general kind of computation requires a Turing Machine for its execution. This implies that a TM process must have a non-predeterminable amount of memory allocated to it at intermediate instants of its execution; i.e. dynamic memory allocation. Many processes encountered in practice are TM processes. The implication for computational practice is that the hardware (CPU) architecture and its operating system must facilitate dynamic memory allocation, and that the programming language used to specify TM processes must have statements with the semantic attribute of dynamic memory allocation, for in Alan Turing"s thesis on computation (1936) the "standard description" of a process is invariant over the most general data that the process is designed to process; i.e. the program describing the process should never have to be modified to allow for differences in the data that is to be processed in different instantiations; i.e. data-invariant programming. Any non-trivial program is partitioned into sub-programs (procedures, subroutines, functions, modules, etc). Examination of the calls/returns between the subprograms reveals that they are nodes in a tree-structure; this tree-structure is independent of the programming language used to encode (define) the process. Each sub-program typically needs some memory for its own use (to store values intermediate between its received data and its computed results); this locally required memory is not needed before the subprogram commences execution, and it is not needed after its execution terminates

  8. High-Risk Breast Lesions: A Machine Learning Model to Predict Pathologic Upgrade and Reduce Unnecessary Surgical Excision.

    PubMed

    Bahl, Manisha; Barzilay, Regina; Yedidia, Adam B; Locascio, Nicholas J; Yu, Lili; Lehman, Constance D

    2018-03-01

    Purpose To develop a machine learning model that allows high-risk breast lesions (HRLs) diagnosed with image-guided needle biopsy that require surgical excision to be distinguished from HRLs that are at low risk for upgrade to cancer at surgery and thus could be surveilled. Materials and Methods Consecutive patients with biopsy-proven HRLs who underwent surgery or at least 2 years of imaging follow-up from June 2006 to April 2015 were identified. A random forest machine learning model was developed to identify HRLs at low risk for upgrade to cancer. Traditional features such as age and HRL histologic results were used in the model, as were text features from the biopsy pathologic report. Results One thousand six HRLs were identified, with a cancer upgrade rate of 11.4% (115 of 1006). A machine learning random forest model was developed with 671 HRLs and tested with an independent set of 335 HRLs. Among the most important traditional features were age and HRL histologic results (eg, atypical ductal hyperplasia). An important text feature from the pathologic reports was "severely atypical." Instead of surgical excision of all HRLs, if those categorized with the model to be at low risk for upgrade were surveilled and the remainder were excised, then 97.4% (37 of 38) of malignancies would have been diagnosed at surgery, and 30.6% (91 of 297) of surgeries of benign lesions could have been avoided. Conclusion This study provides proof of concept that a machine learning model can be applied to predict the risk of upgrade of HRLs to cancer. Use of this model could decrease unnecessary surgery by nearly one-third and could help guide clinical decision making with regard to surveillance versus surgical excision of HRLs. © RSNA, 2017.

  9. Comparison of machine-learning algorithms to build a predictive model for detecting undiagnosed diabetes - ELSA-Brasil: accuracy study.

    PubMed

    Olivera, André Rodrigues; Roesler, Valter; Iochpe, Cirano; Schmidt, Maria Inês; Vigo, Álvaro; Barreto, Sandhi Maria; Duncan, Bruce Bartholow

    2017-01-01

    Type 2 diabetes is a chronic disease associated with a wide range of serious health complications that have a major impact on overall health. The aims here were to develop and validate predictive models for detecting undiagnosed diabetes using data from the Longitudinal Study of Adult Health (ELSA-Brasil) and to compare the performance of different machine-learning algorithms in this task. Comparison of machine-learning algorithms to develop predictive models using data from ELSA-Brasil. After selecting a subset of 27 candidate variables from the literature, models were built and validated in four sequential steps: (i) parameter tuning with tenfold cross-validation, repeated three times; (ii) automatic variable selection using forward selection, a wrapper strategy with four different machine-learning algorithms and tenfold cross-validation (repeated three times), to evaluate each subset of variables; (iii) error estimation of model parameters with tenfold cross-validation, repeated ten times; and (iv) generalization testing on an independent dataset. The models were created with the following machine-learning algorithms: logistic regression, artificial neural network, naïve Bayes, K-nearest neighbor and random forest. The best models were created using artificial neural networks and logistic regression. -These achieved mean areas under the curve of, respectively, 75.24% and 74.98% in the error estimation step and 74.17% and 74.41% in the generalization testing step. Most of the predictive models produced similar results, and demonstrated the feasibility of identifying individuals with highest probability of having undiagnosed diabetes, through easily-obtained clinical data.

  10. Modelling and simulation of effect of ultrasonic vibrations on machining of Ti6Al4V.

    PubMed

    Patil, Sandip; Joshi, Shashikant; Tewari, Asim; Joshi, Suhas S

    2014-02-01

    The titanium alloys cause high machining heat generation and consequent rapid wear of cutting tool edges during machining. The ultrasonic assisted turning (UAT) has been found to be very effective in machining of various materials; especially in the machining of "difficult-to-cut" material like Ti6Al4V. The present work is a comprehensive study involving 2D FE transient simulation of UAT in DEFORM framework and their experimental characterization. The simulation shows that UAT reduces the stress level on cutting tool during machining as compared to that of in continuous turning (CT) barring the penetration stage, wherein both tools are subjected to identical stress levels. There is a 40-45% reduction in cutting forces and about 48% reduction in cutting temperature in UAT over that of in CT. However, the reduction magnitude reduces with an increase in the cutting speed. The experimental analysis of UAT process shows that the surface roughness in UAT is lower than in CT, and the UATed surfaces have matte finish as against the glossy finish on the CTed surfaces. Microstructural observations of the chips and machined surfaces in both processes reveal that the intensity of thermal softening and shear band formation is reduced in UAT over that of in CT. Copyright © 2013 Elsevier B.V. All rights reserved.

  11. Natural Language-based Machine Learning Models for the Annotation of Clinical Radiology Reports.

    PubMed

    Zech, John; Pain, Margaret; Titano, Joseph; Badgeley, Marcus; Schefflein, Javin; Su, Andres; Costa, Anthony; Bederson, Joshua; Lehar, Joseph; Oermann, Eric Karl

    2018-05-01

    Purpose To compare different methods for generating features from radiology reports and to develop a method to automatically identify findings in these reports. Materials and Methods In this study, 96 303 head computed tomography (CT) reports were obtained. The linguistic complexity of these reports was compared with that of alternative corpora. Head CT reports were preprocessed, and machine-analyzable features were constructed by using bag-of-words (BOW), word embedding, and Latent Dirichlet allocation-based approaches. Ultimately, 1004 head CT reports were manually labeled for findings of interest by physicians, and a subset of these were deemed critical findings. Lasso logistic regression was used to train models for physician-assigned labels on 602 of 1004 head CT reports (60%) using the constructed features, and the performance of these models was validated on a held-out 402 of 1004 reports (40%). Models were scored by area under the receiver operating characteristic curve (AUC), and aggregate AUC statistics were reported for (a) all labels, (b) critical labels, and (c) the presence of any critical finding in a report. Sensitivity, specificity, accuracy, and F1 score were reported for the best performing model's (a) predictions of all labels and (b) identification of reports containing critical findings. Results The best-performing model (BOW with unigrams, bigrams, and trigrams plus average word embeddings vector) had a held-out AUC of 0.966 for identifying the presence of any critical head CT finding and an average 0.957 AUC across all head CT findings. Sensitivity and specificity for identifying the presence of any critical finding were 92.59% (175 of 189) and 89.67% (191 of 213), respectively. Average sensitivity and specificity across all findings were 90.25% (1898 of 2103) and 91.72% (18 351 of 20 007), respectively. Simpler BOW methods achieved results competitive with those of more sophisticated approaches, with an average AUC for presence of any

  12. Development of thermal model to analyze thermal flux distribution in thermally enhanced machining of high chrome white cast iron

    NASA Astrophysics Data System (ADS)

    Ravi, A. M.; Murigendrappa, S. M.

    2018-04-01

    In recent times, thermally enhanced machining (TEM) slowly gearing up to cut hard metals like high chrome white cast iron (HCWCI) which were impossible in conventional procedures. Also setting up of suitable cutting parameters and positioning of the heat source against the work appears to be critical in order to enhance the machinability characteristics of the work material. In this research work, the Oxy - LPG flame was used as the heat source and HCWCI as the workpiece. ANSYS-CFD-Flow software was used to develop the transient thermal model to analyze the thermal flux distribution on the work surface during TEM of HCWCI using Cubic boron nitride (CBN) tools. Non-contact type Infrared thermo sensor was used to measure the surface temperature continuously at different positions, and is validated with the thermal model results. The result confirms thermal model is a better predictive tool for thermal flux distribution analysis in TEM process.

  13. Integrated modeling and analysis of the multiple electromechanical couplings for the direct driven feed system in machine tools

    NASA Astrophysics Data System (ADS)

    Yang, Xiaojun; Lu, Dun; Liu, Hui; Zhao, Wanhua

    2018-06-01

    The complicated electromechanical coupling phenomena due to different kinds of causes have significant influences on the dynamic precision of the direct driven feed system in machine tools. In this paper, a novel integrated modeling and analysis method of the multiple electromechanical couplings for the direct driven feed system in machine tools is presented. At first, four different kinds of electromechanical coupling phenomena in the direct driven feed system are analyzed systematically. Then a novel integrated modeling and analysis method of the electromechanical coupling which is influenced by multiple factors is put forward. In addition, the effects of multiple electromechanical couplings on the dynamic precision of the feed system and their main influencing factors are compared and discussed, respectively. Finally, the results of modeling and analysis are verified by the experiments. It finds out that multiple electromechanical coupling loops, which are overlapped and influenced by each other, are the main reasons of the displacement fluctuations in the direct driven feed system.

  14. Prediction of ttt curves of cold working tool steels using support vector machine model

    NASA Astrophysics Data System (ADS)

    Pillai, Nandakumar; Karthikeyan, R., Dr.

    2018-04-01

    The cold working tool steels are of high carbon steels with metallic alloy additions which impart higher hardenability, abrasion resistance and less distortion in quenching. The microstructure changes occurring in tool steel during heat treatment is of very much importance as the final properties of the steel depends upon these changes occurred during the process. In order to obtain the desired performance the alloy constituents and its ratio plays a vital role as the steel transformation itself is complex in nature and depends very much upon the time and temperature. The proper treatment can deliver satisfactory results, at the same time process deviation can completely spoil the results. So knowing time temperature transformation (TTT) of phases is very critical which varies for each type depending upon its constituents and proportion range. To obtain adequate post heat treatment properties the percentage of retained austenite should be lower and metallic carbides obtained should be fine in nature. Support vector machine is a computational model which can learn from the observed data and use these to predict or solve using mathematical model. Back propagation feedback network will be created and trained for further solutions. The points on the TTT curve for the known transformations curves are used to plot the curves for different materials. These data will be trained to predict TTT curves for other steels having similar alloying constituents but with different proportion range. The proposed methodology can be used for prediction of TTT curves for cold working steels and can be used for prediction of phases for different heat treatment methods.

  15. Constant size descriptors for accurate machine learning models of molecular properties

    NASA Astrophysics Data System (ADS)

    Collins, Christopher R.; Gordon, Geoffrey J.; von Lilienfeld, O. Anatole; Yaron, David J.

    2018-06-01

    Two different classes of molecular representations for use in machine learning of thermodynamic and electronic properties are studied. The representations are evaluated by monitoring the performance of linear and kernel ridge regression models on well-studied data sets of small organic molecules. One class of representations studied here counts the occurrence of bonding patterns in the molecule. These require only the connectivity of atoms in the molecule as may be obtained from a line diagram or a SMILES string. The second class utilizes the three-dimensional structure of the molecule. These include the Coulomb matrix and Bag of Bonds, which list the inter-atomic distances present in the molecule, and Encoded Bonds, which encode such lists into a feature vector whose length is independent of molecular size. Encoded Bonds' features introduced here have the advantage of leading to models that may be trained on smaller molecules and then used successfully on larger molecules. A wide range of feature sets are constructed by selecting, at each rank, either a graph or geometry-based feature. Here, rank refers to the number of atoms involved in the feature, e.g., atom counts are rank 1, while Encoded Bonds are rank 2. For atomization energies in the QM7 data set, the best graph-based feature set gives a mean absolute error of 3.4 kcal/mol. Inclusion of 3D geometry substantially enhances the performance, with Encoded Bonds giving 2.4 kcal/mol, when used alone, and 1.19 kcal/mol, when combined with graph features.

  16. Comprehensive modeling of monthly mean soil temperature using multivariate adaptive regression splines and support vector machine

    NASA Astrophysics Data System (ADS)

    Mehdizadeh, Saeid; Behmanesh, Javad; Khalili, Keivan

    2017-07-01

    Soil temperature (T s) and its thermal regime are the most important factors in plant growth, biological activities, and water movement in soil. Due to scarcity of the T s data, estimation of soil temperature is an important issue in different fields of sciences. The main objective of the present study is to investigate the accuracy of multivariate adaptive regression splines (MARS) and support vector machine (SVM) methods for estimating the T s. For this aim, the monthly mean data of the T s (at depths of 5, 10, 50, and 100 cm) and meteorological parameters of 30 synoptic stations in Iran were utilized. To develop the MARS and SVM models, various combinations of minimum, maximum, and mean air temperatures (T min, T max, T); actual and maximum possible sunshine duration; sunshine duration ratio (n, N, n/N); actual, net, and extraterrestrial solar radiation data (R s, R n, R a); precipitation (P); relative humidity (RH); wind speed at 2 m height (u 2); and water vapor pressure (Vp) were used as input variables. Three error statistics including root-mean-square-error (RMSE), mean absolute error (MAE), and determination coefficient (R 2) were used to check the performance of MARS and SVM models. The results indicated that the MARS was superior to the SVM at different depths. In the test and validation phases, the most accurate estimations for the MARS were obtained at the depth of 10 cm for T max, T min, T inputs (RMSE = 0.71 °C, MAE = 0.54 °C, and R 2 = 0.995) and for RH, V p, P, and u 2 inputs (RMSE = 0.80 °C, MAE = 0.61 °C, and R 2 = 0.996), respectively.

  17. Man not a machine: Models, minds, and mental labor, c.1980.

    PubMed

    Stadler, Max

    2017-01-01

    This essay is concerned with the fate of the so-called "computer metaphor" of the mind in the age of mass computing. As such, it is concerned with the ways the mighty metaphor of the rational, rule-based, and serial "information processor," which dominated neurological and psychological theorizing in the early post-WW2 era, came apart during the 1970s and 1980s; and how it was, step by step, replaced by a set of model entities more closely in tune with the significance that was now discerned in certain kinds of "everyday practical action" as the ultimate manifestation of the human mind. By taking a closer look at the ailments and promises of the so-called postindustrial age and more specifically, at the "hazards" associated with the introduction of computers into the workplace, it is shown how models and visions of the mind responded to this new state of affairs. It was in this context-the transformations of mental labor, c.1980-my argument goes, that the minds of men and women revealed themselves to be not so much like computing machines, as the "classic" computer metaphor of the mind, which had birthed the "cognitive revolution" of the 1950s and 1960s, once had it; they were positively unlike them. Instead of "rules" or "symbol manipulation," the minds of computer-equipped brainworkers thus evoked a different set of metaphors: at stake in postindustrial cognition, as this essay argues, was something "parallel," "tacit," and "embodied and embedded." © 2017 Elsevier B.V. All rights reserved.

  18. A Comparison of Costs of Searching the Machine-Readable Data Bases ERIC and "Psychological Abstracts" in an Annual Subscription Rate System Against Costs Estimated for the Same Searches Done in the Lockheed DIALOG System and the System Development Corporation for ERIC, and the Lockheed DIALOG System and PASAT for "Psychological Abstracts."

    ERIC Educational Resources Information Center

    Palmer, Crescentia

    A comparison of costs for computer-based searching of Psychological Abstracts and Educational Resources Information Center (ERIC) systems by the New York State Library at Albany was produced by combining data available from search request forms and from bills from the contract subscription service, the State University of New…

  19. Machine Learning Based Multi-Physical-Model Blending for Enhancing Renewable Energy Forecast -- Improvement via Situation Dependent Error Correction

    SciTech Connect

    Lu, Siyuan; Hwang, Youngdeok; Khabibrakhmanov, Ildar

    With increasing penetration of solar and wind energy to the total energy supply mix, the pressing need for accurate energy forecasting has become well-recognized. Here we report the development of a machine-learning based model blending approach for statistically combining multiple meteorological models for improving the accuracy of solar/wind power forecast. Importantly, we demonstrate that in addition to parameters to be predicted (such as solar irradiance and power), including additional atmospheric state parameters which collectively define weather situations as machine learning input provides further enhanced accuracy for the blended result. Functional analysis of variance shows that the error of individual modelmore » has substantial dependence on the weather situation. The machine-learning approach effectively reduces such situation dependent error thus produces more accurate results compared to conventional multi-model ensemble approaches based on simplistic equally or unequally weighted model averaging. Validation over an extended period of time results show over 30% improvement in solar irradiance/power forecast accuracy compared to forecasts based on the best individual model.« less

  20. Robust Brain-Machine Interface Design Using Optimal Feedback Control Modeling and Adaptive Point Process Filtering

    PubMed Central

    Carmena, Jose M.

    2016-01-01

    Much progress has been made in brain-machine interfaces (BMI) using decoders such as Kalman filters and finding their parameters with closed-loop decoder adaptation (CLDA). However, current decoders do not model the spikes directly, and hence may limit the processing time-scale of BMI control and adaptation. Moreover, while specialized CLDA techniques for intention estimation and assisted training exist, a unified and systematic CLDA framework that generalizes across different setups is lacking. Here we develop a novel closed-loop BMI training architecture that allows for processing, control, and adaptation using spike events, enables robust control and extends to various tasks. Moreover, we develop a unified control-theoretic CLDA framework within which intention estimation, assisted training, and adaptation are performed. The architecture incorporates an infinite-horizon optimal feedback-control (OFC) model of the brain’s behavior in closed-loop BMI control, and a point process model of spikes. The OFC model infers the user’s motor intention during CLDA—a process termed intention estimation. OFC is also used to design an autonomous and dynamic assisted training technique. The point process model allows for neural processing, control and decoder adaptation with every spike event and at a faster time-scale than current decoders; it also enables dynamic spike-event-based parameter adaptation unlike current CLDA methods that use batch-based adaptation on much slower adaptation time-scales. We conducted closed-loop experiments in a non-human primate over tens of days to dissociate the effects of these novel CLDA components. The OFC intention estimation improved BMI performance compared with current intention estimation techniques. OFC assisted training allowed the subject to consistently achieve proficient control. Spike-event-based adaptation resulted in faster and more consistent performance convergence compared with batch-based methods, and was robust to

  1. Exact analytical modeling of magnetic vector potential in surface inset permanent magnet DC machines considering magnet segmentation

    NASA Astrophysics Data System (ADS)

    Jabbari, Ali

    2018-01-01

    Surface inset permanent magnet DC machine can be used as an alternative in automation systems due to their high efficiency and robustness. Magnet segmentation is a common technique in order to mitigate pulsating torque components in permanent magnet machines. An accurate computation of air-gap magnetic field distribution is necessary in order to calculate machine performance. An exact analytical method for magnetic vector potential calculation in surface inset permanent magnet machines considering magnet segmentation has been proposed in this paper. The analytical method is based on the resolution of Laplace and Poisson equations as well as Maxwell equation in polar coordinate by using sub-domain method. One of the main contributions of the paper is to derive an expression for the magnetic vector potential in the segmented PM region by using hyperbolic functions. The developed method is applied on the performance computation of two prototype surface inset magnet segmented motors with open circuit and on load conditions. The results of these models are validated through FEM method.

  2. A modeling of elementary passes taking into account the firing angle in abrasive water jet machining of titanium alloy

    NASA Astrophysics Data System (ADS)

    Bui, Van-Hung; Gilles, Patrick; Cohen, Guillaume; Rubio, Walter

    2018-05-01

    The use of titanium alloys in the aeronautical and high technology domains is widespread. The high strength and the low mass are two outstanding characteristics of titanium alloys which permit to produce parts for these domains. As other hard materials, it is challenging to generate 3D surfaces (e.g. pockets) when using conventional cutting methods. The development of Abrasive Water Jet Machining (AWJM) technology shows the capability to cut any kind of materials and it seems to be a good solution for such titanium materials with low specific force, low deformation of parts and low thermal shocks. Applying this technology for generating 3D surfaces requires to adopt a modelling approach. However, a general methodology results in complex models due to a lot of parameters of the machining process and based on numerous experiments. This study introduces an extended geometry model of an elementary pass when changing the firing angle during machining Ti-6AL-4V titanium alloy with a given machine configuration. Several experiments are conducted to observe the influence of major kinematic operating parameters, i.e. jet inclination angle (α) (perpendicular to the feed direction) and traverse speed (Vf). The material exposure time and the erosion capability of abrasives particles are affected directly by a variation of the traverse speed (Vf) and firing angle (α). These variations lead to different erosion rates along the kerf profile characterized by the depth and width of cut. A comparison demonstrated an efficiency of the proposed model for depth and width of elementary passes. Based on knowledge of the influence of both firing angle and traverse speed on the elementary pass shape, the proposed model allows to develop the simulation of AWJM process and paves a way for milling flat bottom pockets and 3D complex shapes.

  3. A comparative study of machine learning methods for time-to-event survival data for radiomics risk modelling.

    PubMed

    Leger, Stefan; Zwanenburg, Alex; Pilz, Karoline; Lohaus, Fabian; Linge, Annett; Zöphel, Klaus; Kotzerke, Jörg; Schreiber, Andreas; Tinhofer, Inge; Budach, Volker; Sak, Ali; Stuschke, Martin; Balermpas, Panagiotis; Rödel, Claus; Ganswindt, Ute; Belka, Claus; Pigorsch, Steffi; Combs, Stephanie E; Mönnich, David; Zips, Daniel; Krause, Mechthild; Baumann, Michael; Troost, Esther G C; Löck, Steffen; Richter, Christian

    2017-10-16

    Radiomics applies machine learning algorithms to quantitative imaging data to characterise the tumour phenotype and predict clinical outcome. For the development of radiomics risk models, a variety of different algorithms is available and it is not clear which one gives optimal results. Therefore, we assessed the performance of 11 machine learning algorithms combined with 12 feature selection methods by the concordance index (C-Index), to predict loco-regional tumour control (LRC) and overall survival for patients with head and neck squamous cell carcinoma. The considered algorithms are able to deal with continuous time-to-event survival data. Feature selection and model building were performed on a multicentre cohort (213 patients) and validated using an independent cohort (80 patients). We found several combinations of machine learning algorithms and feature selection methods which achieve similar results, e.g. C-Index = 0.71 and BT-COX: C-Index = 0.70 in combination with Spearman feature selection. Using the best performing models, patients were stratified into groups of low and high risk of recurrence. Significant differences in LRC were obtained between both groups on the validation cohort. Based on the presented analysis, we identified a subset of algorithms which should be considered in future radiomics studies to develop stable and clinically relevant predictive models for time-to-event endpoints.

  4. Global-scale assessment of groundwater depletion and related groundwater abstractions: Combining hydrological modeling with information from well observations and GRACE satellites

    NASA Astrophysics Data System (ADS)

    Döll, Petra; Müller Schmied, Hannes; Schuh, Carina; Portmann, Felix T.; Eicker, Annette

    2014-07-01

    Groundwater depletion (GWD) compromises crop production in major global agricultural areas and has negative ecological consequences. To derive GWD at the grid cell, country, and global levels, we applied a new version of the global hydrological model WaterGAP that simulates not only net groundwater abstractions and groundwater recharge from soils but also groundwater recharge from surface water bodies in dry regions. A large number of independent estimates of GWD as well as total water storage (TWS) trends determined from GRACE satellite data by three analysis centers were compared to model results. GWD and TWS trends are simulated best assuming that farmers in GWD areas irrigate at 70% of optimal water requirement. India, United States, Iran, Saudi Arabia, and China had the highest GWD rates in the first decade of the 21st century. On the Arabian Peninsula, in Libya, Egypt, Mali, Mozambique, and Mongolia, at least 30% of the abstracted groundwater was taken from nonrenewable groundwater during this time period. The rate of global GWD has likely more than doubled since the period 1960-2000. Estimated GWD of 113 km3/yr during 2000-2009, corresponding to a sea level rise of 0.31 mm/yr, is much smaller than most previous estimates. About 15% of the globally abstracted groundwater was taken from nonrenewable groundwater during this period. To monitor recent temporal dynamics of GWD and related water abstractions, GRACE data are best evaluated with a hydrological model that, like WaterGAP, simulates the impact of abstractions on water storage, but the low spatial resolution of GRACE remains a challenge.

  5. The Aachen miniaturized heart-lung machine--first results in a small animal model.

    PubMed

    Schnoering, Heike; Arens, Jutta; Sachweh, Joerg S; Veerman, Melanie; Tolba, Rene; Schmitz-Rode, Thomas; Steinseifer, Ulrich; Vazquez-Jimenez, Jaime F

    2009-11-01

    Congenital heart surgery most often incorporates extracorporeal circulation. Due to foreign surface contact and the administration of foreign blood in many children, inflammatory response and hemolysis are important matters of debate. This is particularly an issue in premature and low birth-weight newborns. Taking these considerations into account, the Aachen miniaturized heart-lung machine (MiniHLM) with a total static priming volume of 102 mL (including tubing) was developed and tested in a small animal model. Fourteen female Chinchilla Bastard rabbits were operated on using two different kinds of circuits. In eight animals, a conventional HLM with Dideco Kids oxygenator and Stöckert roller pump (Sorin group, Milan, Italy) was used, and the Aachen MiniHLM was employed in six animals. Outcome parameters were hemolysis and blood gas analysis including lactate. The rabbits were anesthetized, and a standard median sternotomy was performed. The ascending aorta and the right atrium were cannulated. After initiating cardiopulmonary bypass, the aorta was cross-clamped, and cardiac arrest was induced by blood cardioplegia. Blood samples for hemolysis and blood gas analysis were drawn before, during, and after cardiopulmonary bypass. After 1 h aortic clamp time, all animals were weaned from cardiopulmonary bypass. Blood gas analysis revealed adequate oxygenation and perfusion during cardiopulmonary bypass, irrespective of the employed perfusion system. The use of the Aachen MiniHLM resulted in a statistically significant reduced decrease in fibrinogen during cardiopulmonary bypass. A trend revealing a reduced increase in free hemoglobin during bypass in the MiniHLM group could also be observed. This newly developed Aachen MiniHLM with low priming volume, reduced hemolysis, and excellent gas transfer (O(2) and CO(2)) may reduce circuit-induced complications during heart surgery in neonates.

  6. Fluid-structure interaction modeling of wind turbines: simulating the full machine

    NASA Astrophysics Data System (ADS)

    Hsu, Ming-Chen; Bazilevs, Yuri

    2012-12-01

    In this paper we present our aerodynamics and fluid-structure interaction (FSI) computational techniques that enable dynamic, fully coupled, 3D FSI simulation of wind turbines at full scale, and in the presence of the nacelle and tower (i.e., simulation of the "full machine"). For the interaction of wind and flexible blades we employ a nonmatching interface discretization approach, where the aerodynamics is computed using a low-order finite-element-based ALE-VMS technique, while the rotor blades are modeled as thin composite shells discretized using NURBS-based isogeometric analysis (IGA). We find that coupling FEM and IGA in this manner gives a good combination of efficiency, accuracy, and flexibility of the computational procedures for wind turbine FSI. The interaction between the rotor and tower is handled using a non-overlapping sliding-interface approach, where both moving- and stationary-domain formulations of aerodynamics are employed. At the fluid-structure and sliding interfaces, the kinematic and traction continuity is enforced weakly, which is a key ingredient of the proposed numerical methodology. We present several simulations of a three-blade 5~MW wind turbine, with and without the tower. We find that, in the case of no tower, the presence of the sliding interface has no effect on the prediction of aerodynamic loads on the rotor. From this we conclude that weak enforcement of the kinematics gives just as accurate results as the strong enforcement, and thus enables the simulation of rotor-tower interaction (as well as other applications involving mechanical components in relative motion). We also find that the blade passing the tower produces a 10-12 % drop (per blade) in the aerodynamic torque. We feel this finding may be important when it comes to the fatigue-life analysis and prediction for wind turbine blades.

  7. Neural control and adaptive neural forward models for insect-like, energy-efficient, and adaptable locomotion of walking machines

    PubMed Central

    Manoonpong, Poramate; Parlitz, Ulrich; Wörgötter, Florentin

    2013-01-01

    Living creatures, like walking animals, have found fascinating solutions for the problem of locomotion control. Their movements show the impression of elegance including versatile, energy-efficient, and adaptable locomotion. During the last few decades, roboticists have tried to imitate such natural properties with artificial legged locomotion systems by using different approaches including machine learning algorithms, classical engineering control techniques, and biologically-inspired control mechanisms. However, their levels of performance are still far from the natural ones. By contrast, animal locomotion mechanisms seem to largely depend not only on central mechanisms (central pattern generators, CPGs) and sensory feedback (afferent-based control) but also on internal forward models (efference copies). They are used to a different degree in different animals. Generally, CPGs organize basic rhythmic motions which are shaped by sensory feedback while internal models are used for sensory prediction and state estimations. According to this concept, we present here adaptive neural locomotion control consisting of a CPG mechanism with neuromodulation and local leg control mechanisms based on sensory feedback and adaptive neural forward models with efference copies. This neural closed-loop controller enables a walking machine to perform a multitude of different walking patterns including insect-like leg movements and gaits as well as energy-efficient locomotion. In addition, the forward models allow the machine to autonomously adapt its locomotion to deal with a change of terrain, losing of ground contact during stance phase, stepping on or hitting an obstacle during swing phase, leg damage, and even to promote cockroach-like climbing behavior. Thus, the results presented here show that the employed embodied neural closed-loop system can be a powerful way for developing robust and adaptable machines. PMID:23408775

  8. A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification.

    PubMed

    Wen, Cuihong; Zhang, Jing; Rebelo, Ana; Cheng, Fanyong

    2016-01-01

    Optical Music Recognition (OMR) has received increasing attention in recent years. In this paper, we propose a classifier based on a new method named Directed Acyclic Graph-Large margin Distribution Machine (DAG-LDM). The DAG-LDM is an improvement of the Large margin Distribution Machine (LDM), which is a binary classifier that optimizes the margin distribution by maximizing the margin mean and minimizing the margin variance simultaneously. We modify the LDM to the DAG-LDM to solve the multi-class music symbol classification problem. Tests are conducted on more than 10000 music symbol images, obtained from handwritten and printed images of music scores. The proposed method provides superior classification capability and achieves much higher classification accuracy than the state-of-the-art algorithms such as Support Vector Machines (SVMs) and Neural Networks (NNs).

  9. A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification

    PubMed Central

    Wen, Cuihong; Zhang, Jing; Rebelo, Ana; Cheng, Fanyong

    2016-01-01

    Optical Music Recognition (OMR) has received increasing attention in recent years. In this paper, we propose a classifier based on a new method named Directed Acyclic Graph-Large margin Distribution Machine (DAG-LDM). The DAG-LDM is an improvement of the Large margin Distribution Machine (LDM), which is a binary classifier that optimizes the margin distribution by maximizing the margin mean and minimizing the margin variance simultaneously. We modify the LDM to the DAG-LDM to solve the multi-class music symbol classification problem. Tests are conducted on more than 10000 music symbol images, obtained from handwritten and printed images of music scores. The proposed method provides superior classification capability and achieves much higher classification accuracy than the state-of-the-art algorithms such as Support Vector Machines (SVMs) and Neural Networks (NNs). PMID:26985826

  10. Coupling surface water (Delft3D) to groundwater (MODFLOW) in the Bay-Delta community model: the effect of major abstractions in the Delta

    NASA Astrophysics Data System (ADS)

    Hendriks, D.; Ball, S. M.; Van der Wegen, M.; Verkaik, J.; van Dam, A.

    2016-12-01

    We present a coupled groundwater-surface water model for the San Francisco Bay and Sacramento Valley that consists of a combination of a spatially-distributed groundwater model (Modflow) based on the USGS Central Valley model(1) and the Flexible Mesh (FM) surface water model of the Bay Area(2). With this coupled groundwater-surface water model, we assessed effects of climate, surface water abstractions and groundwater pumping on surface water and groundwater levels, groundwater-surface water interaction and infiltration/seepage fluxes. Results show that the effect of climate (high flow and low flow) on surface water and groundwater is significant and most prominent in upstream areas. The surface water abstractions cause significant local surface water levels decrease (over 2 m), which may cause inflow of bay water during low flow periods, resulting in salinization of surface water in more upstream areas. Groundwater level drawdown due to surface water withdrawal is moderate and limited to the area of the withdrawals. The groundwater pumping causes large groundwater level drawdowns (up to 0.8 m) and significant changes in seepage/infiltration fluxes in the model. However, the effect on groundwater-surface water exchange is relatively small. The presented model instrument gives a sound first impression of the effects of climate and water abstraction on both surface water and groundwater. The combination of Modflow and Flexible Mesh has potential for modelling of groundwater-surface water exchange in deltaic areas, also in other parts of the world. However, various improvements need to be made in order to make the simulation results useful in practice. In addition, a water quality aspect could be added to assess salinization processes as well as groundwater-surface water aspects of water and soil pollution. (1) http://ca.water.usgs.gov/projects/central-valley/central-valley-hydrologic-model.html (2) www.d3d-baydelta.org

  11. Volumetric error modeling, identification and compensation based on screw theory for a large multi-axis propeller-measuring machine

    NASA Astrophysics Data System (ADS)

    Zhong, Xuemin; Liu, Hongqi; Mao, Xinyong; Li, Bin; He, Songping; Peng, Fangyu

    2018-05-01

    Large multi-axis propeller-measuring machines have two types of geometric error, position-independent geometric errors (PIGEs) and position-dependent geometric errors (PDGEs), which both have significant effects on the volumetric error of the measuring tool relative to the worktable. This paper focuses on modeling, identifying and compensating for the volumetric error of the measuring machine. A volumetric error model in the base coordinate system is established based on screw theory considering all the geometric errors. In order to fully identify all the geometric error parameters, a new method for systematic measurement and identification is proposed. All the PIGEs of adjacent axes and the six PDGEs of the linear axes are identified with a laser tracker using the proposed model. Finally, a volumetric error compensation strategy is presented and an inverse kinematic solution for compensation is proposed. The final measuring and compensation experiments have further verified the efficiency and effectiveness of the measuring and identification method, indicating that the method can be used in volumetric error compensation for large machine tools.

  12. Parameter Estimation of the Thermal Network Model of a Machine Tool Spindle by Self-made Bluetooth Temperature Sensor Module

    PubMed Central

    Lo, Yuan-Chieh; Hu, Yuh-Chung; Chang, Pei-Zen

    2018-01-01

    Thermal characteristic analysis is essential for machine tool spindles because sudden failures may occur due to unexpected thermal issue. This article presents a lumped-parameter Thermal Network Model (TNM) and its parameter estimation scheme, including hardware and software, in order to characterize both the steady-state and transient thermal behavior of machine tool spindles. For the hardware, the authors develop a Bluetooth Temperature Sensor Module (BTSM) which accompanying with three types of temperature-sensing probes (magnetic, screw, and probe). Its specification, through experimental test, achieves to the precision ±(0.1 + 0.0029|t|) °C, resolution 0.00489 °C, power consumption 7 mW, and size Ø40 mm × 27 mm. For the software, the heat transfer characteristics of the machine tool spindle correlative to rotating speed are derived based on the theory of heat transfer and empirical formula. The predictive TNM of spindles was developed by grey-box estimation and experimental results. Even under such complicated operating conditions as various speeds and different initial conditions, the experiments validate that the present modeling methodology provides a robust and reliable tool for the temperature prediction with normalized mean square error of 99.5% agreement, and the present approach is transferable to the other spindles with a similar structure. For realizing the edge computing in smart manufacturing, a reduced-order TNM is constructed by Model Order Reduction (MOR) technique and implemented into the real-time embedded system. PMID:29473877

  13. Parameter Estimation of the Thermal Network Model of a Machine Tool Spindle by Self-made Bluetooth Temperature Sensor Module.

    PubMed

    Lo, Yuan-Chieh; Hu, Yuh-Chung; Chang, Pei-Zen

    2018-02-23

    Thermal characteristic analysis is essential for machine tool spindles because sudden failures may occur due to unexpected thermal issue. This article presents a lumped-parameter Thermal Network Model (TNM) and its parameter estimation scheme, including hardware and software, in order to characterize both the steady-state and transient thermal behavior of machine tool spindles. For the hardware, the authors develop a Bluetooth Temperature Sensor Module (BTSM) which accompanying with three types of temperature-sensing probes (magnetic, screw, and probe). Its specification, through experimental test, achieves to the precision ±(0.1 + 0.0029|t|) °C, resolution 0.00489 °C, power consumption 7 mW, and size Ø40 mm × 27 mm. For the software, the heat transfer characteristics of the machine tool spindle correlative to rotating speed are derived based on the theory of heat transfer and empirical formula. The predictive TNM of spindles was developed by grey-box estimation and experimental results. Even under such complicated operating conditions as various speeds and different initial conditions, the experiments validate that the present modeling methodology provides a robust and reliable tool for the temperature prediction with normalized mean square error of 99.5% agreement, and the present approach is transferable to the other spindles with a similar structure. For realizing the edge computing in smart manufacturing, a reduced-order TNM is constructed by Model Order Reduction (MOR) technique and implemented into the real-time embedded system.

  14. Nanocomposites for Machining Tools

    PubMed Central

    Loginov, Pavel; Mishnaevsky, Leon; Levashov, Evgeny

    2017-01-01

    Machining tools are used in many areas of production. To a considerable extent, the performance characteristics of the tools determine the quality and cost of obtained products. The main materials used for producing machining tools are steel, cemented carbides, ceramics and superhard materials. A promising way to improve the performance characteristics of these materials is to design new nanocomposites based on them. The application of micromechanical modeling during the elaboration of composite materials for machining tools can reduce the financial and time costs for development of new tools, with enhanced performance. This article reviews the main groups of nanocomposites for machining tools and their performance. PMID:29027926

  15. Metaheuristic and Machine Learning Models for TFE-731-2, PW4056, and JT8D-9 Cruise Thrust

    NASA Astrophysics Data System (ADS)

    Baklacioglu, Tolga

    2017-08-01

    The requirement for an accurate engine thrust model has a major antecedence in airline fuel saving programs, assessment of environmental effects of fuel consumption, emissions reduction studies, and air traffic management applications. In this study, utilizing engine manufacturers' real data, a metaheuristic model based on genetic algorithms (GAs) and a machine learning model based on neural networks (NNs) trained with Levenberg-Marquardt (LM), delta-bar-delta (DBD), and conjugate gradient (CG) algorithms were accomplished to incorporate the effect of both flight altitude and Mach number in the estimation of thrust. For the GA model, the analysis of population size impact on the model's accuracy and effect of number of data on model coefficients were also performed. For the NN model, design of optimum topology was searched for one- and two-hidden-layer networks. Predicted thrust values presented a close agreement with real thrust data for both models, among which LM trained NNs gave the best accuracies.

  16. The Development of a Machine Learning Inpatient Acute Kidney Injury Prediction Model.

    PubMed

    Koyner, Jay L; Carey, Kyle A; Edelson, Dana P; Churpek, Matthew M

    2018-07-01

    To develop an acute kidney injury risk prediction model using electronic health record data for longitudinal use in hospitalized patients. Observational cohort study. Tertiary, urban, academic medical center from November 2008 to January 2016. All adult inpatients without pre-existing renal failure at admission, defined as first serum creatinine greater than or equal to 3.0 mg/dL, International Classification of Diseases, 9th Edition, code for chronic kidney disease stage 4 or higher or having received renal replacement therapy within 48 hours of first serum creatinine measurement. None. Demographics, vital signs, diagnostics, and interventions were used in a Gradient Boosting Machine algorithm to predict serum creatinine-based Kidney Disease Improving Global Outcomes stage 2 acute kidney injury, with 60% of the data used for derivation and 40% for validation. Area under the receiver operator characteristic curve (AUC) was calculated in the validation cohort, and subgroup analyses were conducted across admission serum creatinine, acute kidney injury severity, and hospital location. Among the 121,158 included patients, 17,482 (14.4%) developed any Kidney Disease Improving Global Outcomes acute kidney injury, with 4,251 (3.5%) developing stage 2. The AUC (95% CI) was 0.90 (0.90-0.90) for predicting stage 2 acute kidney injury within 24 hours and 0.87 (0.87-0.87) within 48 hours. The AUC was 0.96 (0.96-0.96) for receipt of renal replacement therapy (n = 821) in the next 48 hours. Accuracy was similar across hospital settings (ICU, wards, and emergency department) and admitting serum creatinine groupings. At a probability threshold of greater than or equal to 0.022, the algorithm had a sensitivity of 84% and a specificity of 85% for stage 2 acute kidney injury and predicted the development of stage 2 a median of 41 hours (interquartile range, 12-141 hr) prior to the development of stage 2 acute kidney injury. Readily available electronic health record data can be

  17. Extracting aerobic system dynamics during unsupervised activities of daily living using wearable sensor machine learning models.

    PubMed

    Beltrame, Thomas; Amelard, Robert; Wong, Alexander; Hughson, Richard L

    2018-02-01

    . This study is the first to use wearable sensors in unsupervised activities of daily living in combination with novel machine learning algorithms to investigate the aerobic system dynamics with the potential to contribute to models of functional health status and guide future individualized health care in the normal population.

  18. Solving the AI Planning Plus Scheduling Problem Using Model Checking via Automatic Translation from the Abstract Plan Preparation Language (APPL) to the Symbolic Analysis Laboratory (SAL)

    NASA Technical Reports Server (NTRS)

    Butler, Ricky W.; Munoz, Cesar A.; Siminiceanu, Radu I.

    2007-01-01

    This paper describes a translator from a new planning language named the Abstract Plan Preparation Language (APPL) to the Symbolic Analysis Laboratory (SAL) model checker. This translator has been developed in support of the Spacecraft Autonomy for Vehicles and Habitats (SAVH) project sponsored by the Exploration Technology Development Program, which is seeking to mature autonomy technology for the vehicles and operations centers of Project Constellation.

  19. An asymptotical machine

    NASA Astrophysics Data System (ADS)

    Cristallini, Achille

    2016-07-01

    A new and intriguing machine may be obtained replacing the moving pulley of a gun tackle with a fixed point in the rope. Its most important feature is the asymptotic efficiency. Here we obtain a satisfactory description of this machine by means of vector calculus and elementary trigonometry. The mathematical model has been compared with experimental data and briefly discussed.

  20. Abstracts and program proceedings of the 1994 meeting of the International Society for Ecological Modelling North American Chapter

    SciTech Connect

    Kercher, J.R.

    1994-06-01

    This document contains information about the 1994 meeting of the International Society for Ecological Modelling North American Chapter. The topics discussed include: extinction risk assessment modelling, ecological risk analysis of uranium mining, impacts of pesticides, demography, habitats, atmospheric deposition, and climate change.

  1. MOAtox: A comprehensive mode of action and acute aquatic toxicity database for predictive model development (SETAC abstract)

    EPA Science Inventory

    The mode of toxic action (MOA) has been recognized as a key determinant of chemical toxicity and as an alternative to chemical class-based predictive toxicity modeling. However, the development of quantitative structure activity relationship (QSAR) and other models has been limit...

  2. (abstract) A Test of the Theoretical Models of Bipolar Outflows: The Bipolar Outflow in Mon R2

    NASA Technical Reports Server (NTRS)

    Xie, Taoling; Goldsmith, Paul; Patel, Nimesh

    1993-01-01

    We report some results of a study of the massive bipolar outflow in the central region of the relatively nearby giant molecular cloud Monoceros R2. We make a quantative comparison of our results with the Shu et al. outflow model which incorporates a radially directed wind sweeping up the ambient material into a shell. We find that this simple model naturally explains the shape of this thin shell. Although Shu's model in its simplest form predicts with reasonable parameters too much mass at very small polar angles, as previously pointed out by Masson and Chernin, it provides a reasonable good fit to the mass distribution at larger polar angles. It is possible that this discrepancy is due to inhomogeneities of the ambient molecular gas which is not considered by the model. We also discuss the constraints imposed by these results on recent jet-driven outflow models.

  3. Analytical Model-Based Design Optimization of a Transverse Flux Machine

    SciTech Connect

    Hasan, Iftekhar; Husain, Tausif; Sozer, Yilmaz

    This paper proposes an analytical machine design tool using magnetic equivalent circuit (MEC)-based particle swarm optimization (PSO) for a double-sided, flux-concentrating transverse flux machine (TFM). The magnetic equivalent circuit method is applied to analytically establish the relationship between the design objective and the input variables of prospective TFM designs. This is computationally less intensive and more time efficient than finite element solvers. A PSO algorithm is then used to design a machine with the highest torque density within the specified power range along with some geometric design constraints. The stator pole length, magnet length, and rotor thickness are the variablesmore » that define the optimization search space. Finite element analysis (FEA) was carried out to verify the performance of the MEC-PSO optimized machine. The proposed analytical design tool helps save computation time by at least 50% when compared to commercial FEA-based optimization programs, with results found to be in agreement with less than 5% error.« less

  4. A hybrid machine learning model to estimate nitrate contamination of production zone groundwater in the Central Valley, California

    NASA Astrophysics Data System (ADS)

    Ransom, K.; Nolan, B. T.; Faunt, C. C.; Bell, A.; Gronberg, J.; Traum, J.; Wheeler, D. C.; Rosecrans, C.; Belitz, K.; Eberts, S.; Harter, T.

    2016-12-01

    A hybrid, non-linear, machine learning statistical model was developed within a statistical learning framework to predict nitrate contamination of groundwater to depths of approximately 500 m below ground surface in the Central Valley, California. A database of 213 predictor variables representing well characteristics, historical and current field and county scale nitrogen mass balance, historical and current landuse, oxidation/reduction conditions, groundwater flow, climate, soil characteristics, depth to groundwater, and groundwater age were assigned to over 6,000 private supply and public supply wells measured previously for nitrate and located throughout the study area. The machine learning method, gradient boosting machine (GBM) was used to screen predictor variables and rank them in order of importance in relation to the groundwater nitrate measurements. The top five most important predictor variables included oxidation/reduction characteristics, historical field scale nitrogen mass balance, climate, and depth to 60 year old water. Twenty-two variables were selected for the final model and final model errors for log-transformed hold-out data were R squared of 0.45 and root mean square error (RMSE) of 1.124. Modeled mean groundwater age was tested separately for error improvement in the model and when included decreased model RMSE by 0.5% compared to the same model without age and by 0.20% compared to the model with all 213 variables. 1D and 2D partial plots were examined to determine how variables behave individually and interact in the model. Some variables behaved as expected: log nitrate decreased with increasing probability of anoxic conditions and depth to 60 year old water, generally decreased with increasing natural landuse surrounding wells and increasing mean groundwater age, generally increased with increased minimum depth to high water table and with increased base flow index value. Other variables exhibited much more erratic or noisy behavior in

  5. Modeling the stress dependence of Barkhausen phenomena for stress axis linear and noncollinear with applied magnetic field (abstract)

    SciTech Connect

    Sablik, M.J.; Augustyniak, B.; Chmielewski, M.

    1996-04-01

    The almost linear dependence of the maximum Barkhausen noise signal amplitude on stress has made it a tool for nondestructive evaluation of residual stress. Recently, a model has been developed to account for the stress dependence of the Barkhausen noise signal. The model uses the development of Alessandro {ital et} {ital al}. who use coupled Langevin equations to derive an expression for the Barkhausen noise power spectrum. The model joins this expression to the magnetomechanical hysteresis model of Sablik {ital et} {ital al}., obtaining both a hysteretic and stress-dependent result for the magnetic-field-dependent Barkhausen noise envelope and obtaining specifically themore » almost linear stress dependence of the Barkhausen noise maximum experimentally. In this paper, we extend the model to derive the angular dependence observed by Kwun of the Barkhausen noise amplitude when stress axis is taken at different angles relative to magnetic field. We also apply the model to the experimental observation that in XC10 French steel, there is an apparent almost linear correlation with stress of hysteresis loss and of the integral of the Barkhausen noise signal over applied field {ital H}. Further, the two quantities, Barkhausen noise integral and hysteresis loss, are linearly correlated with each other. The model shows how that behavior is to be expected for the measured steel because of its sharply rising hysteresis curve. {copyright} {ital 1996 American Institute of Physics.}« less

  6. Machine learning-based kinetic modeling: a robust and reproducible solution for quantitative analysis of dynamic PET data.

    PubMed

    Pan, Leyun; Cheng, Caixia; Haberkorn, Uwe; Dimitrakopoulou-Strauss, Antonia

    2017-05-07

    A variety of compartment models are used for the quantitative analysis of dynamic positron emission tomography (PET) data. Traditionally, these models use an iterative fitting (IF) method to find the least squares between the measured and calculated values over time, which may encounter some problems such as the overfitting of model parameters and a lack of reproducibility, especially when handling noisy data or error data. In this paper, a machine learning (ML) based kinetic modeling method is introduced, which can fully utilize a historical reference database to build a moderate kinetic model directly dealing with noisy data but not trying to smooth the noise in the image. Also, due to the database, the presented method is capable of automatically adjusting the models using a multi-thread grid parameter searching technique. Furthermore, a candidate competition concept is proposed to combine the advantages of the ML and IF modeling methods, which could find a balance between fitting to historical data and to the unseen target curve. The machine learning based method provides a robust and reproducible solution that is user-independent for VOI-based and pixel-wise quantitative analysis of dynamic PET data.

  7. Using Perturbed Physics Ensembles and Machine Learning to Select Parameters for Reducing Regional Biases in a Global Climate Model

    NASA Astrophysics Data System (ADS)

    Li, S.; Rupp, D. E.; Hawkins, L.; Mote, P.; McNeall, D. J.; Sarah, S.; Wallom, D.; Betts, R. A.

    2017-12-01

    This study investigates the potential to reduce known summer hot/dry biases over Pacific Northwest in the UK Met Office's atmospheric model (HadAM3P) by simultaneously varying multiple model parameters. The bias-reduction process is done through a series of steps: 1) Generation of perturbed physics ensemble (PPE) through the volunteer computing network weather@home; 2) Using machine learning to train "cheap" and fast statistical emulators of climate model, to rule out regions of parameter spaces that lead to model variants that do not satisfy observational constraints, where the observational constraints (e.g., top-of-atmosphere energy flux, magnitude of annual temperature cycle, summer/winter temperature and precipitation) are introduced sequentially; 3) Designing a new PPE by "pre-filtering" using the emulator results. Steps 1) through 3) are repeated until results are considered to be satisfactory (3 times in our case). The process includes a sensitivity analysis to find dominant parameters for various model output metrics, which reduces the number of parameters to be perturbed with each new PPE. Relative to observational uncertainty, we achieve regional improvements without introducing large biases in other parts of the globe. Our results illustrate the potential of using machine learning to train cheap and fast statistical emulators of climate model, in combination with PPEs in systematic model improvement.

  8. Machine learning-based kinetic modeling: a robust and reproducible solution for quantitative analysis of dynamic PET data

    NASA Astrophysics Data System (ADS)

    Pan, Leyun; Cheng, Caixia; Haberkorn, Uwe; Dimitrakopoulou-Strauss, Antonia

    2017-05-01

    A variety of compartment models are used for the quantitative analysis of dynamic positron emission tomography (PET) data. Traditionally, these models use an iterative fitting (IF) method to find the least squares between the measured and calculated values over time, which may encounter some problems such as the overfitting of model parameters and a lack of reproducibility, especially when handling noisy data or error data. In this paper, a machine learning (ML) based kinetic modeling method is introduced, which can fully utilize a historical reference database to build a moderate kinetic model directly dealing with noisy data but not trying to smooth the noise in the image. Also, due to the database, the presented method is capable of automatically adjusting the models using a multi-thread grid parameter searching technique. Furthermore, a candidate competition concept is proposed to combine the advantages of the ML and IF modeling methods, which could find a balance between fitting to historical data and to the unseen target curve. The machine learning based method provides a robust and reproducible solution that is user-independent for VOI-based and pixel-wise quantitative analysis of dynamic PET data.

  9. The Abstraction Process of Limit Knowledge

    ERIC Educational Resources Information Center

    Sezgin Memnun, Dilek; Aydin, Bünyamin; Özbilen, Ömer; Erdogan, Günes

    2017-01-01

    The RBC+C abstraction model is an effective model in mathematics education because it gives the opportunity to analyze research data through cognitive actions. For this reason, we aim to examine the abstraction process of the limit knowledge of two volunteer participant students using the RBC+C abstraction model. With this aim, the students'…

  10. The evolving market structures of gambling: case studies modelling the socioeconomic assignment of gaming machines in Melbourne and Sydney, Australia.

    PubMed

    Marshall, David C; Baker, Robert G V

    2002-01-01

    The expansion of gambling industries worldwide is intertwined with the growing government dependence on gambling revenue for fiscal assignments. In Australia, electronic gaming machines (EGMs) have dominated recent gambling industry growth. As EGMs have proliferated, growing recognition has emerged that EGM distribution closely reflects levels of socioeconomic disadvantage. More machines are located in less advantaged regions. This paper analyses time-series socioeconomic distributions of EGMs in Melbourne, Australia, an immature EGM market, and then compares the findings with the mature market in Sydney. Similar findings in both cities suggest that market assignment of EGMs transcends differences in historical and legislative environments. This indicates that similar underlying structures are evident in both markets. Modelling the spatial structures of gambling markets provides an opportunity to identify regions most at risk of gambling related problems. Subsequently, policies can be formulated which ensure fiscal revenue from gambling can be better targeted towards regions likely to be most afflicted by excessive gambling-related problems.

  11. ABSTRACTS OF RESEARCH REPORTS.

    DTIC Science & Technology

    DENTISTRY, ABSTRACTS), TEETH, DISEASES, MOUTH, TRANSPLANTATION, HYGIENE, STERILIZATION, FLUORIDES, HISTOLOGY, SURGICAL IMPLANTATION, OXYTETRACYCLINE , GELATINS, CELLULOSE, CASTING, PROGRAMMED INSTRUCTION, TRAINING DEVICES

  12. Abstracting Concepts and Methods.

    ERIC Educational Resources Information Center

    Borko, Harold; Bernier, Charles L.

    This text provides a complete discussion of abstracts--their history, production, organization, publication--and of indexing. Instructions for abstracting are outlined, and standards and criteria for abstracting are stated. Management, automation, and personnel are discussed in terms of possible economies that can be derived from the introduction…

  13. Abstraction and Consolidation

    ERIC Educational Resources Information Center

    Monaghan, John; Ozmantar, Mehmet Fatih

    2006-01-01

    The framework for this paper is a recently developed theory of abstraction in context. The paper reports on data collected from one student working on tasks concerned with absolute value functions. It examines the relationship between mathematical constructions and abstractions. It argues that an abstraction is a consolidated construction that can…

  14. A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

    PubMed

    Allyn, Jérôme; Allou, Nicolas; Augustin, Pascal; Philip, Ivan; Martinet, Olivier; Belghiti, Myriem; Provenchere, Sophie; Montravers, Philippe; Ferdynus, Cyril

    2017-01-01

    The benefits of cardiac surgery are sometimes difficult to predict and the decision to operate on a given individual is complex. Machine Learning and Decision Curve Analysis (DCA) are recent methods developed to create and evaluate prediction models. We conducted a retrospective cohort study using a prospective collected database from December 2005 to December 2012, from a cardiac surgical center at University Hospital. The different models of prediction of mortality in-hospital after elective cardiac surgery, including EuroSCORE II, a logistic regression model and a machine learning model, were compared by ROC and DCA. Of the 6,520 patients having elective cardiac surgery with cardiopulmonary bypass, 6.3% died. Mean age was 63.4 years old (standard deviation 14.4), and mean EuroSCORE II was 3.7 (4.8) %. The area under ROC curve (IC95%) for the machine learning model (0.795 (0.755-0.834)) was significantly higher than EuroSCORE II or the logistic regression model (respectively, 0.737 (0.691-0.783) and 0.742 (0.698-0.785), p < 0.0001). Decision Curve Analysis showed that the machine learning model, in this monocentric study, has a greater benefit whatever the probability threshold. According to ROC and DCA, machine learning model is more accurate in predicting mortality after elective cardiac surgery than EuroSCORE II. These results confirm the use of machine learning methods in the field of medical prediction.

  15. A model composition for Mars derived from the oxygen isotopic ratios of martian/SNC meteorites. [Abstract only

    NASA Technical Reports Server (NTRS)

    Delaney, J. S.

    1994-01-01

    Oxygen is the most abundant element in most meteorites, yet the ratios of its isotopes are seldom used to constrain the compositional history of achondrites. The two major achondrite groups have O isotope signatures that differ from any plausible chondritic precursors and lie between the ordinary and carbonaceous chondrite domains. If the assumption is made that the present global sampling of chondritic meteorites reflects the variability of O reservoirs at the time of planetessimal/planet aggregation in the early nebula, then the O in these groups must reflect mixing between known chondritic reservoirs. This approach, in combination with constraints based on Fe-Mn-Mg systematics, has been used previously to model the composition of the basaltic achondrite parent body (BAP) and provides a model precursor composition that is generally consistent with previous eucrite parent body (EPB) estimates. The same approach is applied to Mars exploiting the assumption that the SNC and related meteorites sample the martian lithosphere. Model planet and planetesimal compositions can be derived by mixing of known chondritic components using O isotope ratios as the fundamental compositional constraint. The major- and minor-element composition for Mars derived here and that derived previously for the basaltic achondrite parent body are, in many respects, compatible with model compositions generated using completely independent constraints. The role of volatile elements and alkalis in particular remains a major difficulty in applying such models.

  16. Validation of TGLF in C-Mod and DIII-D using machine learning and integrated modeling tools

    NASA Astrophysics Data System (ADS)

    Rodriguez-Fernandez, P.; White, Ae; Cao, Nm; Creely, Aj; Greenwald, Mj; Grierson, Ba; Howard, Nt; Meneghini, O.; Petty, Cc; Rice, Je; Sciortino, F.; Yuan, X.

    2017-10-01

    Predictive models for steady-state and perturbative transport are necessary to support burning plasma operations. A combination of machine learning algorithms and integrated modeling tools is used to validate TGLF in C-Mod and DIII-D. First, a new code suite, VITALS, is used to compare SAT1 and SAT0 models in C-Mod. VITALS exploits machine learning and optimization algorithms for the validation of transport codes. Unlike SAT0, the SAT1 saturation rule contains a model to capture cross-scale turbulence coupling. Results show that SAT1 agrees better with experiments, further confirming that multi-scale effects are needed to model heat transport in C-Mod L-modes. VITALS will next be used to analyze past data from DIII-D: L-mode ``Shortfall'' plasma and ECH swing experiments. A second code suite, PRIMA, allows for integrated modeling of the plasma response to Laser Blow-Off cold pulses. Preliminary results show that SAT1 qualitatively reproduces the propagation of cold pulses after LBO injections and SAT0 does not, indicating that cross-scale coupling effects play a role in the plasma response. PRIMA will be used to ``predict-first'' cold pulse experiments using the new LBO system at DIII-D, and analyze existing ECH heat pulse data. Work supported by DE-FC02-99ER54512, DE-FC02-04ER54698.

  17. Weibull Multiplicative Model and Machine Learning Models for Full-Automatic Dark-Spot Detection from SAR Images

    NASA Astrophysics Data System (ADS)

    Taravat, A.; Del Frate, F.

    2013-09-01

    As a major aspect of marine pollution, oil release into the sea has serious biological and environmental impacts. Among remote sensing systems (which is a tool that offers a non-destructive investigation method), synthetic aperture radar (SAR) can provide valuable synoptic information about the position and size of the oil spill due to its wide area coverage and day/night, and all-weather capabilities. In this paper we present a new automated method for oil-spill monitoring. A new approach is based on the combination of Weibull Multiplicative Model and machine learning techniques to differentiate between dark spots and the background. First, the filter created based on Weibull Multiplicative Model is applied to each sub-image. Second, the sub-image is segmented by two different neural networks techniques (Pulsed Coupled Neural Networks and Multilayer Perceptron Neural Networks). As the last step, a very simple filtering process is used to eliminate the false targets. The proposed approaches were tested on 20 ENVISAT and ERS2 images which contained dark spots. The same parameters were used in all tests. For the overall dataset, the average accuracies of 94.05 % and 95.20 % were obtained for PCNN and MLP methods, respectively. The average computational time for dark-spot detection with a 256 × 256 image in about 4 s for PCNN segmentation using IDL software which is the fastest one in this field at present. Our experimental results demonstrate that the proposed approach is very fast, robust and effective. The proposed approach can be applied to the future spaceborne SAR images.

  18. Statistical Mechanics of Coherent Ising Machine — The Case of Ferromagnetic and Finite-Loading Hopfield Models

    NASA Astrophysics Data System (ADS)

    Aonishi, Toru; Mimura, Kazushi; Utsunomiya, Shoko; Okada, Masato; Yamamoto, Yoshihisa

    2017-10-01

    The coherent Ising machine (CIM) has attracted attention as one of the most effective Ising computing architectures for solving large scale optimization problems because of its scalability and high-speed computational ability. However, it is difficult to implement the Ising computation in the CIM because the theories and techniques of classical thermodynamic equilibrium Ising spin systems cannot be directly applied to the CIM. This means we have to adapt these theories and techniques to the CIM. Here we focus on a ferromagnetic model and a finite loading Hopfield model, which are canonical models sharing a common mathematical structure with almost all other Ising models. We derive macroscopic equations to capture nonequilibrium phase transitions in these models. The statistical mechanical methods developed here constitute a basis for constructing evaluation methods for other Ising computation models.

  19. A Bayesian network model for predicting aquatic toxicity mode of action using two dimensional theoretical molecular descriptors-abstract

    EPA Science Inventory

    The mode of toxic action (MoA) has been recognized as a key determinant of chemical toxicity but MoA classification in aquatic toxicology has been limited. We developed a Bayesian network model to classify aquatic toxicity mode of action using a recently published dataset contain...

  20. Modeling of industrial stream and resources of machine-building enterpriser complex of wood preparation

    NASA Astrophysics Data System (ADS)

    Sereda, T. G.; Kostarev, S. N.

    2018-03-01

    Theoretical bases of linkage of material streams of the machine-building enterprise and the automated system of decision-making are developed. The process of machine-building manufacture is submitted by the existential system. The equation of preservation of movement is based on calculation of volume of manufacture. The basis of resource variables includes capacities and operators of the equipment. Indignations such as a defect and failure are investigated in the existential basis. The equation of a stream of details on a manufacturing route is made. The received analytical expression expresses a condition of a stream of movement of details in view of influence of work of the equipment and traumatism of the personnel.

  1. Predictive modeling for corrective maintenance of imaging devices from machine logs.

    PubMed

    Patil, Ravindra B; Patil, Meru A; Ravi, Vidya; Naik, Sarif

    2017-07-01

    In the cost sensitive healthcare industry, an unplanned downtime of diagnostic and therapy imaging devices can be a burden on the financials of both the hospitals as well as the original equipment manufacturers (OEMs). In the current era of connectivity, it is easier to get these devices connected to a standard monitoring station. Once the system is connected, OEMs can monitor the health of these devices remotely and take corrective actions by providing preventive maintenance thereby avoiding major unplanned downtime. In this article, we present an overall methodology of predicting failure of these devices well before customer experiences it. We use data-driven approach based on machine learning to predict failures in turn resulting in reduced machine downtime, improved customer satisfaction and cost savings for the OEMs. One of the use-case of predicting component failure of PHILIPS iXR system is explained in this article.

  2. Characterizing the SEMG patterns with myofascial pain using a multi-scale wavelet model through machine learning approaches.

    PubMed

    Lin, Yu-Ching; Yu, Nan-Ying; Jiang, Ching-Fen; Chang, Shao-Hsia

    2018-06-02

    In this paper, we introduce a newly developed multi-scale wavelet model for the interpretation of surface electromyography (SEMG) signals and validate the model's capability to characterize changes in neuromuscular activation in cases with myofascial pain syndrome (MPS) via machine learning methods. The SEMG data collected from normal (N = 30; 27 women, 3 men) and MPS subjects (N = 26; 22 women, 4 men) were adopted for this retrospective analysis. SMEGs were measured from the taut-band loci on both sides of the trapezius muscle on the upper back while he/she conducted a cyclic bilateral backward shoulder extension movement within 1 min. Classification accuracy of the SEMG model to differentiate MPS patients from normal subjects was 77% using template matching and 60% using K-means clustering. Classification consistency between the two machine learning methods was 87% in the normal group and 93% in the MPS group. The 2D feature graphs derived from the proposed multi-scale model revealed distinct patterns between normal subjects and MPS patients. The classification consistency using template matching and K-means clustering suggests the pote