The Experts below are selected from a list of 96 Experts worldwide ranked by ideXlab platform

J. Callahan - One of the best experts on this subject based on the ideXlab platform.

  • Towards Developing Verifiable Neural Network Controller
    1996
    Co-Authors: Wu Wen, J. Callahan, Marcello Napolitano
    Abstract:

    Artificial Neural Networks(ANN) play an important part in developing intelligent and autonomous systems. By training the ANN with desired input-output patterns that are derived from human experience or simulation, the designers of such system avoid specifying detailed analytical model of such complex systems. Unlike conventional system design and development which is based on an iterative procedure of system analysis, specification, implementation and testing, ANN based system relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper, we propose the Neuralware engineering framework to address the above issues. It is based on an iterative approach on specification, model checking, and testing. After the ANN-based system is designed and trained u..

  • Neuralware Engineering: Develop Verifiable ANN-based Systems
    1996
    Co-Authors: Wu Wen, J. Callahan
    Abstract:

    Artificial Neural Networks(ANN) play an important part in developing intelligent robotic and autonomous systems. By training the ANN with desired input-output patterns that are derived from human experience or simulation, the designers of such system avoid specifying detailed analytical model of such complex systems. Unlike conventional system design and development which is based on an iterative procedure of system analysis, specification, implementation and testing, ANN based system relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper, we propose the Neuralware engineering framework to address the above issues. This framework is based on our experience with verifying and testing complex software systems such as communication protocols and..

  • Neuralware engineering: develop verifiable ANN-based systems
    Proceedings IEEE International Joint Symposia on Intelligence and Systems, 1
    Co-Authors: Wu Wen, J. Callahan
    Abstract:

    Artificial neural networks (ANN) play an important part in developing intelligent robotic and autonomous systems; it relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper we propose the Neuralware engineering framework to address the above issues. This framework is based on our experience with verifying and testing complex software systems. It is based on an iterative approach on specification, model checking, and testing. After the ANN-based system is designed and trained using an initial partial system model, a rule extraction algorithm is used to discover what has been learned. The discrepancies between the learned rules and the model is compared to modify the system model. This process is repeated until the behavior of the real system is validated against the model and specification.

Wu Wen - One of the best experts on this subject based on the ideXlab platform.

  • Towards Developing Verifiable Neural Network Controller
    1996
    Co-Authors: Wu Wen, J. Callahan, Marcello Napolitano
    Abstract:

    Artificial Neural Networks(ANN) play an important part in developing intelligent and autonomous systems. By training the ANN with desired input-output patterns that are derived from human experience or simulation, the designers of such system avoid specifying detailed analytical model of such complex systems. Unlike conventional system design and development which is based on an iterative procedure of system analysis, specification, implementation and testing, ANN based system relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper, we propose the Neuralware engineering framework to address the above issues. It is based on an iterative approach on specification, model checking, and testing. After the ANN-based system is designed and trained u..

  • Neuralware Engineering: Develop Verifiable ANN-based Systems
    1996
    Co-Authors: Wu Wen, J. Callahan
    Abstract:

    Artificial Neural Networks(ANN) play an important part in developing intelligent robotic and autonomous systems. By training the ANN with desired input-output patterns that are derived from human experience or simulation, the designers of such system avoid specifying detailed analytical model of such complex systems. Unlike conventional system design and development which is based on an iterative procedure of system analysis, specification, implementation and testing, ANN based system relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper, we propose the Neuralware engineering framework to address the above issues. This framework is based on our experience with verifying and testing complex software systems such as communication protocols and..

  • Neuralware engineering: develop verifiable ANN-based systems
    Proceedings IEEE International Joint Symposia on Intelligence and Systems, 1
    Co-Authors: Wu Wen, J. Callahan
    Abstract:

    Artificial neural networks (ANN) play an important part in developing intelligent robotic and autonomous systems; it relies on training to formulate the control mechanisms. When such ANN-based components are embedded in a larger system, their interactions become harder to analyze and model. Formal testing of such system for safety properties is extremely hard due to the lack of a complete system model. In this paper we propose the Neuralware engineering framework to address the above issues. This framework is based on our experience with verifying and testing complex software systems. It is based on an iterative approach on specification, model checking, and testing. After the ANN-based system is designed and trained using an initial partial system model, a rule extraction algorithm is used to discover what has been learned. The discrepancies between the learned rules and the model is compared to modify the system model. This process is repeated until the behavior of the real system is validated against the model and specification.

Richard S. Segall - One of the best experts on this subject based on the ideXlab platform.

  • Review of data, text and web mining software
    Kybernetes, 2010
    Co-Authors: Qingyu Zhang, Richard S. Segall
    Abstract:

    Purpose – The purpose of this paper is to review and compare selected software for data mining, text mining (TM), and web mining that are not available as free open‐source software.Design/methodology/approach – Selected softwares are compared with their common and unique features. The software for data mining are SAS® Enterprise Miner™, Megaputer PolyAnalyst® 5.0, Neuralware Predict®, and BioDiscovery GeneSight®. The software for TM are CompareSuite, SAS® Text Miner, TextAnalyst, VisualText, Megaputer PolyAnalyst® 5.0, and WordStat. The software for web mining are Megaputer PolyAnalyst®, SPSS Clementine®, ClickTracks, and QL2.Findings – This paper discusses and compares the existing features, characteristics, and algorithms of selected software for data mining, TM, and web mining, respectively. These softwares are also applied to available data sets.Research limitations/implications – The limitations are the inclusion of selected software and datasets rather than considering the entire realm of these. Thi...

  • Comparing Four-Selected Data Mining Software
    Software Applications, 2009
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    This chapter discusses four-selected software for data mining that are not available as free opensource software. The four-selected software for data mining are SAS® Enterprise MinerTM, Megaputer PolyAnalyst® 5.0, Neuralware Predict® and BioDiscovery GeneSight®, each of which was provided by partnerships with our university. These software are described and compared by their existing features, characteristics, and algorithms and also applied to a large database of forest cover types with 63,377 rows and 54 attributes. Background on related literature and software are also presented. Screen shots of each of the four-selected software are presented, as are future directions and conclusions.

  • 2006a), Applications of neural network and genetic algorithm data mining techniques in bioinformatics knowledge discovery – A preliminary study
    2008
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    This paper presents preliminary research in the area of the applications of modern heuristics and data mining techniques in knowledge discovery. Specifically applications of data mining for neural networks using Neuralware Predict ® software and genetic algorithms using Biodiscovery GeneSight ® software were selected for bioscience data sets of continuous numerical valued abalone fish data. Conclusions and future directions of the research are presented. Large volume of data and complexity in problem solving inspire research in data mining and modern heuristics. Data mining (i.e., knowledge discovery) is the process of automating information discovery. It is the process of analyzing data from different perspectives, summarizing it into useful information, and finding different patterns (e.g., classification, regression, and clustering). Many problems are difficult t

  • Data visualization and data mining of continuous numerical and discrete nominal‐valued microarray databases for bioinformatics
    Kybernetes, 2006
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    Purpose – To present research in the area of the applications of modern heuristics and data mining techniques in knowledge discovery.Design/methodology/approach – Applications of data mining for neural networks using Neuralware Predict® software, genetic algorithms using Biodiscovery GeneSight® (2005) software, and regression and discriminant analysis using SPSS® were selected for bioscience data sets of continuous numerical‐valued Abalone fish data and discrete nominal‐valued mushroom data.Findings – This paper illustrates the useful information that can be obtained using data mining for evolutionary algorithms specifically as those for neural networks, genetic algorithms, regression analysis, and discriminant analysis.Research limitations/implications – The use of Neuralware Predict® was a very effective method of implementing training rules for neural networks to identify the important attributes of numerical and nominal valued data.Practical implications – The software and algorithms discussed in the ...

  • Encyclopedia of Data Warehousing and Mining - Comparing Four-Selected Data Mining Software
    Encyclopedia of Data Warehousing and Mining Second Edition, 1
    Co-Authors: Richard S. Segall
    Abstract:

    This chapter discusses four-selected software for data mining that are not available as free open-source software. The four-selected software for data mining are SAS® Enterprise MinerTM, Megaputer PolyAnalyst® 5.0, Neuralware Predict® and BioDiscovery GeneSight ®, each of which was provided by partnerships with our university. These software are described and compared by their existing features, characteristics, and algorithms and also applied to a large database of forest cover types with 63,377 rows and 54 attributes. Background on related literature and software are also presented. Screen shots of each of the four-selected software are presented, as are future directions and conclusions.

Qingyu Zhang - One of the best experts on this subject based on the ideXlab platform.

  • Review of data, text and web mining software
    Kybernetes, 2010
    Co-Authors: Qingyu Zhang, Richard S. Segall
    Abstract:

    Purpose – The purpose of this paper is to review and compare selected software for data mining, text mining (TM), and web mining that are not available as free open‐source software.Design/methodology/approach – Selected softwares are compared with their common and unique features. The software for data mining are SAS® Enterprise Miner™, Megaputer PolyAnalyst® 5.0, Neuralware Predict®, and BioDiscovery GeneSight®. The software for TM are CompareSuite, SAS® Text Miner, TextAnalyst, VisualText, Megaputer PolyAnalyst® 5.0, and WordStat. The software for web mining are Megaputer PolyAnalyst®, SPSS Clementine®, ClickTracks, and QL2.Findings – This paper discusses and compares the existing features, characteristics, and algorithms of selected software for data mining, TM, and web mining, respectively. These softwares are also applied to available data sets.Research limitations/implications – The limitations are the inclusion of selected software and datasets rather than considering the entire realm of these. Thi...

  • Comparing Four-Selected Data Mining Software
    Software Applications, 2009
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    This chapter discusses four-selected software for data mining that are not available as free opensource software. The four-selected software for data mining are SAS® Enterprise MinerTM, Megaputer PolyAnalyst® 5.0, Neuralware Predict® and BioDiscovery GeneSight®, each of which was provided by partnerships with our university. These software are described and compared by their existing features, characteristics, and algorithms and also applied to a large database of forest cover types with 63,377 rows and 54 attributes. Background on related literature and software are also presented. Screen shots of each of the four-selected software are presented, as are future directions and conclusions.

  • 2006a), Applications of neural network and genetic algorithm data mining techniques in bioinformatics knowledge discovery – A preliminary study
    2008
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    This paper presents preliminary research in the area of the applications of modern heuristics and data mining techniques in knowledge discovery. Specifically applications of data mining for neural networks using Neuralware Predict ® software and genetic algorithms using Biodiscovery GeneSight ® software were selected for bioscience data sets of continuous numerical valued abalone fish data. Conclusions and future directions of the research are presented. Large volume of data and complexity in problem solving inspire research in data mining and modern heuristics. Data mining (i.e., knowledge discovery) is the process of automating information discovery. It is the process of analyzing data from different perspectives, summarizing it into useful information, and finding different patterns (e.g., classification, regression, and clustering). Many problems are difficult t

  • Data visualization and data mining of continuous numerical and discrete nominal‐valued microarray databases for bioinformatics
    Kybernetes, 2006
    Co-Authors: Richard S. Segall, Qingyu Zhang
    Abstract:

    Purpose – To present research in the area of the applications of modern heuristics and data mining techniques in knowledge discovery.Design/methodology/approach – Applications of data mining for neural networks using Neuralware Predict® software, genetic algorithms using Biodiscovery GeneSight® (2005) software, and regression and discriminant analysis using SPSS® were selected for bioscience data sets of continuous numerical‐valued Abalone fish data and discrete nominal‐valued mushroom data.Findings – This paper illustrates the useful information that can be obtained using data mining for evolutionary algorithms specifically as those for neural networks, genetic algorithms, regression analysis, and discriminant analysis.Research limitations/implications – The use of Neuralware Predict® was a very effective method of implementing training rules for neural networks to identify the important attributes of numerical and nominal valued data.Practical implications – The software and algorithms discussed in the ...

  • Using Data Mining for Forecasting Data Management Needs
    Handbook of Computational Intelligence in Manufacturing and Production Management, 1
    Co-Authors: Qingyu Zhang, Richard S. Segall
    Abstract:

    This chapter illustrates the use of data mining as a computational intelligence methodology for forecasting data management needs. Specifically, this chapter discusses the use of data mining with multidimensional databases for determining data management needs for the selected biotechnology data of forest cover data (63,377 rows and 54 attributes) and human lung cancer data set (12,600 rows of transcript sequences and 156 columns of gene types). The data mining is performed using four selected software of SAS® Enterprise MinerTM, Megaputer PolyAnalyst® 5.0, Neuralware Predict®, and Bio- Discovery GeneSight®. The analysis and results will be used to enhance the intelligence capabilities of biotechnology research by improving data visualization and forecasting for organizations. The tools and techniques discussed here can be representative of those applicable in a typical manufacturing and production environment. Screen shots of each of the four selected software are presented, as are conclusions and future directions.

Nancy J. Minshew - One of the best experts on this subject based on the ideXlab platform.

  • Dynamic cortical systems subserving cognition: fMRI studies with typical and atypical individuals
    Erlbaum, 2001
    Co-Authors: Patricia A. Carpenter, Marcel Adam Just, Timothy A. Keller, Vladimir Cherkassky, Jennifer K. Roth, Nancy J. Minshew
    Abstract:

    Functional brain imaging brings many new offerings to the table of cognitive neuroscience, particularly offerings that help refine our understanding of the dynamic and adaptive properties of brain function that underpin learning and development. It is a little ironic that a methodology that is classically associated with static, still images, should be informative about dynamics and plasticity. But the association is no longer correct. In the early days of functional neuroimaging, in the late 1980’s, the scanners were less sensitive, so that all that was possible was a group average image depicting the areas of brain activation, averaged over several participants, contrasting a small number of experimental conditions. The current instrumentation, namely high-speed fMRI, is far more sensitive yielding enough data to observe reliable effects in single participants in a few minutes per experimental condition. Capturing the dynamics requires only a fast enough shutter speed and a willing subject. One of the consequences of the new technology of particular relevance here is that it is possible to examine the cognitive system as it adapts to slight differences in the quantitative and qualitative demands of a particular experimental condition. The result is that the brain adaptation is clearly manifest, and this manifestation is one of the key offerings of fMRI: the hardware-software distinction in the analysis of cognition that was previously a useful heuristic scientific strategy obscures one of the main adaptive properties of mind. The hardware (brain tissue) is dynamically recruited to meet the processing needs. The underlying software and the Neuralware in whic

  • Dynamic Cortical Systems Subserving Cognition: fMRI Studies With Typical and Atypical Individuals
    2001
    Co-Authors: Patricia A. Carpenter, Marcel Adam Just, Timothy A. Keller, Vladimir Cherkassky, Jennifer Roth, Nancy J. Minshew
    Abstract:

    Functional brain imaging brings many new offerings to the table of cognitive neuroscience, particularly offerings that help refine our understanding of the dynamic and adaptive properties of brain function that underpin learning and development. It is a little ironic that a methodology that is classically associated with static, still images, should be informative about dynamics and plasticity. But the association is no longer correct. In the early days of functional neuroimaging, in the late 1980’s, the scanners were less sensitive, so that all that was possible was a group average image depicting the areas of brain activation, averaged over several participants, contrasting a small number of experimental conditions. The current instrumentation, namely high-speed fMRI, is far more sensitive yielding enough data to observe reliable effects in single participants in a few minutes per experimental condition. Capturing the dynamics requires only a fast enough shutter speed and a willing subject. One of the consequences of the new technology of particular relevance here is that it is possible to examine the cognitive system as it adapts to slight differences in the quantitative and qualitative demands of a particular experimental condition. The result is that the brain adaptation is clearly manifest, and this manifestation is one of the key offerings of fMRI: the hardware-software distinction in the analysis of cognition that was previously a useful heuristic scientific strategy obscures one of the main adaptive properties of mind. The hardware (brain tissue) is dynamically recruited to meet the processing needs. The underlying software and the Neuralware in which it is implemented is constantly changing. It is likely that this dynamic recruitment scheme underlies performance not only in a changing cognitive task, but also underlies adaptation over a longer time frame in a changing world. In this chapter, we explore several implications of a new perspective on cognition and its adaptiveness offered by cognitive brain imaging, focusing on examples in the of area of language comprehension. This chapter does not present the details of the theory, which appears elsewhere (Just, Carpenter & Varma, 1999). Here we summarize some of the main points of this theoretical perspective, and below we examine how they apply to issues of brain plasticity and atypical development. Some of the key hypotheses that are emerging from this perspective include the following: • Cognition entails physiological work and resource consumption. • A system involves collaboration among multiple neural components, a team. • The team members may have multiple and overlapping functions • The components are dynamically recruited • As a behavior becomes more skilled, there is better coordination of its components • Disturbance, such as cortical injury, can result in re-balancing the work load among the components • Developmental syndromes, such as autism, may result in an unusual collaboration, with differential amounts of coordination within vs. among system components We will first describe evidence for these hypotheses based on fMRI studies of normal young adults in highlevel cognitive tasks. Second, we will describe some initial fMRI research with adults who have experienced stroke. Last, we will briefly describe some preliminary fMRI studies of individuals who are high-functioning intellectually, but also have autism.