Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks

David Cornforth

    Research output: Book chapter/Published conference paperConference paper

    Abstract

    How can cerebellar model neural networks be successfully applied to automated classification? Simple modifications and a new training scheme are applied to the Cerebellar Model Articulation Controller (CMAC). This results in a classifier with fast training time, guaranteed convergence and a small number of parameters. How can model parameters be set to achieve optimum classifier accuracy, without lengthy empirical trials? A consideration of the most significant sources of classification error results in a simple method for estimating the optimum range for the parameters. The method is tested using empirical trials, and shown to be reliable. This makes the modified CMAC a desirable choice for automated classification in the context of pattern recognition and data mining.
    Original languageEnglish
    Title of host publicationArtificial Neural Networks and Expert Systems
    EditorsNikola Kasabov, Brendon Woodford
    Place of PublicationDunedin, New Zealand
    PublisherUniversity of Otago
    Pages1-6
    Number of pages6
    ISBN (Electronic)1877139408
    Publication statusPublished - 2001
    EventConference on Artificial Neural Networks and Expert systems - Dunedin, New Zealand, New Zealand
    Duration: 22 Nov 200124 Nov 2001

    Conference

    ConferenceConference on Artificial Neural Networks and Expert systems
    CountryNew Zealand
    Period22/11/0124/11/01

    Fingerprint

    data mining
    pattern recognition
    parameter
    method
    trial

    Cite this

    Cornforth, D. (2001). Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks. In N. Kasabov, & B. Woodford (Eds.), Artificial Neural Networks and Expert Systems (pp. 1-6). Dunedin, New Zealand: University of Otago.
    Cornforth, David. / Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks. Artificial Neural Networks and Expert Systems. editor / Nikola Kasabov ; Brendon Woodford. Dunedin, New Zealand : University of Otago, 2001. pp. 1-6
    @inproceedings{e6ea2afe236d4aa193776b9217efc1b9,
    title = "Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks",
    abstract = "How can cerebellar model neural networks be successfully applied to automated classification? Simple modifications and a new training scheme are applied to the Cerebellar Model Articulation Controller (CMAC). This results in a classifier with fast training time, guaranteed convergence and a small number of parameters. How can model parameters be set to achieve optimum classifier accuracy, without lengthy empirical trials? A consideration of the most significant sources of classification error results in a simple method for estimating the optimum range for the parameters. The method is tested using empirical trials, and shown to be reliable. This makes the modified CMAC a desirable choice for automated classification in the context of pattern recognition and data mining.",
    author = "David Cornforth",
    note = "Imported on 03 May 2017 - DigiTool details were: publisher = Dunedin, New Zealand: University of Otago, 2001. editor/s (773b) = Kasabov, Nikola & Woodford, Brendon; Event dates (773o) = 22-24 November 2001; Parent title (773t) = Conference on Artificial Neural Networks and Expert systems.",
    year = "2001",
    language = "English",
    pages = "1--6",
    editor = "Nikola Kasabov and Brendon Woodford",
    booktitle = "Artificial Neural Networks and Expert Systems",
    publisher = "University of Otago",

    }

    Cornforth, D 2001, Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks. in N Kasabov & B Woodford (eds), Artificial Neural Networks and Expert Systems. University of Otago, Dunedin, New Zealand, pp. 1-6, Conference on Artificial Neural Networks and Expert systems, New Zealand, 22/11/01.

    Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks. / Cornforth, David.

    Artificial Neural Networks and Expert Systems. ed. / Nikola Kasabov; Brendon Woodford. Dunedin, New Zealand : University of Otago, 2001. p. 1-6.

    Research output: Book chapter/Published conference paperConference paper

    TY - GEN

    T1 - Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks

    AU - Cornforth, David

    N1 - Imported on 03 May 2017 - DigiTool details were: publisher = Dunedin, New Zealand: University of Otago, 2001. editor/s (773b) = Kasabov, Nikola & Woodford, Brendon; Event dates (773o) = 22-24 November 2001; Parent title (773t) = Conference on Artificial Neural Networks and Expert systems.

    PY - 2001

    Y1 - 2001

    N2 - How can cerebellar model neural networks be successfully applied to automated classification? Simple modifications and a new training scheme are applied to the Cerebellar Model Articulation Controller (CMAC). This results in a classifier with fast training time, guaranteed convergence and a small number of parameters. How can model parameters be set to achieve optimum classifier accuracy, without lengthy empirical trials? A consideration of the most significant sources of classification error results in a simple method for estimating the optimum range for the parameters. The method is tested using empirical trials, and shown to be reliable. This makes the modified CMAC a desirable choice for automated classification in the context of pattern recognition and data mining.

    AB - How can cerebellar model neural networks be successfully applied to automated classification? Simple modifications and a new training scheme are applied to the Cerebellar Model Articulation Controller (CMAC). This results in a classifier with fast training time, guaranteed convergence and a small number of parameters. How can model parameters be set to achieve optimum classifier accuracy, without lengthy empirical trials? A consideration of the most significant sources of classification error results in a simple method for estimating the optimum range for the parameters. The method is tested using empirical trials, and shown to be reliable. This makes the modified CMAC a desirable choice for automated classification in the context of pattern recognition and data mining.

    M3 - Conference paper

    SP - 1

    EP - 6

    BT - Artificial Neural Networks and Expert Systems

    A2 - Kasabov, Nikola

    A2 - Woodford, Brendon

    PB - University of Otago

    CY - Dunedin, New Zealand

    ER -

    Cornforth D. Building Practical Classifiers Using Cerebellar Model Associative Memory Neural Networks. In Kasabov N, Woodford B, editors, Artificial Neural Networks and Expert Systems. Dunedin, New Zealand: University of Otago. 2001. p. 1-6