A parsimonious radial basis function-based neural network for data classification

Shing Chiang Tan, Chee Peng Lim, Junzo Watada

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    The radial basis function neural network trained with a dynamic decay adjustment (known as RBFNDDA) algorithm exhibits a greedy insertion behavior as a result of recruiting many hidden nodes for encoding information during its training process. In this chapter, a new variant RBFNDDA is proposed to rectify such deficiency. Specifically, the hidden nodes of RBFNDDA are re-organized through the supervised Fuzzy ARTMAP (FAM) classifier, and the parameters of these nodes are adapted using the Harmonic Means (HM) algorithm. The performance of the proposed model is evaluated empirically using three benchmark data sets. The results indicate that the proposed model is able to produce a compact network structure and, at the same time, to provide high classification performances.

    Original languageEnglish
    Title of host publicationSmart Innovation, Systems and Technologies
    PublisherSpringer Science and Business Media Deutschland GmbH
    Pages49-60
    Number of pages12
    Volume42
    ISBN (Print)9783319212081
    DOIs
    Publication statusPublished - 2016
    Event5th International Conference on Intelligent Decision Technologies, 2013 - Sesimbra, Portugal
    Duration: 2013 Jun 262013 Jun 28

    Publication series

    NameSmart Innovation, Systems and Technologies
    Volume42
    ISSN (Print)21903018
    ISSN (Electronic)21903026

    Other

    Other5th International Conference on Intelligent Decision Technologies, 2013
    CountryPortugal
    CitySesimbra
    Period13/6/2613/6/28

    Fingerprint

    Neural networks
    Classifiers
    Radial basis function
    Node
    Decay
    Benchmark
    Recruiting
    Network structure
    Classifier

    Keywords

    • Adaptive resonance theory
    • Classification
    • Harmonic mean algorithm
    • Radial basis function neural network

    ASJC Scopus subject areas

    • Computer Science(all)
    • Decision Sciences(all)

    Cite this

    Tan, S. C., Lim, C. P., & Watada, J. (2016). A parsimonious radial basis function-based neural network for data classification. In Smart Innovation, Systems and Technologies (Vol. 42, pp. 49-60). (Smart Innovation, Systems and Technologies; Vol. 42). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-21209-8_4

    A parsimonious radial basis function-based neural network for data classification. / Tan, Shing Chiang; Lim, Chee Peng; Watada, Junzo.

    Smart Innovation, Systems and Technologies. Vol. 42 Springer Science and Business Media Deutschland GmbH, 2016. p. 49-60 (Smart Innovation, Systems and Technologies; Vol. 42).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Tan, SC, Lim, CP & Watada, J 2016, A parsimonious radial basis function-based neural network for data classification. in Smart Innovation, Systems and Technologies. vol. 42, Smart Innovation, Systems and Technologies, vol. 42, Springer Science and Business Media Deutschland GmbH, pp. 49-60, 5th International Conference on Intelligent Decision Technologies, 2013, Sesimbra, Portugal, 13/6/26. https://doi.org/10.1007/978-3-319-21209-8_4
    Tan SC, Lim CP, Watada J. A parsimonious radial basis function-based neural network for data classification. In Smart Innovation, Systems and Technologies. Vol. 42. Springer Science and Business Media Deutschland GmbH. 2016. p. 49-60. (Smart Innovation, Systems and Technologies). https://doi.org/10.1007/978-3-319-21209-8_4
    Tan, Shing Chiang ; Lim, Chee Peng ; Watada, Junzo. / A parsimonious radial basis function-based neural network for data classification. Smart Innovation, Systems and Technologies. Vol. 42 Springer Science and Business Media Deutschland GmbH, 2016. pp. 49-60 (Smart Innovation, Systems and Technologies).
    @inproceedings{c960930eec4d4918a08139a49d93035e,
    title = "A parsimonious radial basis function-based neural network for data classification",
    abstract = "The radial basis function neural network trained with a dynamic decay adjustment (known as RBFNDDA) algorithm exhibits a greedy insertion behavior as a result of recruiting many hidden nodes for encoding information during its training process. In this chapter, a new variant RBFNDDA is proposed to rectify such deficiency. Specifically, the hidden nodes of RBFNDDA are re-organized through the supervised Fuzzy ARTMAP (FAM) classifier, and the parameters of these nodes are adapted using the Harmonic Means (HM) algorithm. The performance of the proposed model is evaluated empirically using three benchmark data sets. The results indicate that the proposed model is able to produce a compact network structure and, at the same time, to provide high classification performances.",
    keywords = "Adaptive resonance theory, Classification, Harmonic mean algorithm, Radial basis function neural network",
    author = "Tan, {Shing Chiang} and Lim, {Chee Peng} and Junzo Watada",
    year = "2016",
    doi = "10.1007/978-3-319-21209-8_4",
    language = "English",
    isbn = "9783319212081",
    volume = "42",
    series = "Smart Innovation, Systems and Technologies",
    publisher = "Springer Science and Business Media Deutschland GmbH",
    pages = "49--60",
    booktitle = "Smart Innovation, Systems and Technologies",

    }

    TY - GEN

    T1 - A parsimonious radial basis function-based neural network for data classification

    AU - Tan, Shing Chiang

    AU - Lim, Chee Peng

    AU - Watada, Junzo

    PY - 2016

    Y1 - 2016

    N2 - The radial basis function neural network trained with a dynamic decay adjustment (known as RBFNDDA) algorithm exhibits a greedy insertion behavior as a result of recruiting many hidden nodes for encoding information during its training process. In this chapter, a new variant RBFNDDA is proposed to rectify such deficiency. Specifically, the hidden nodes of RBFNDDA are re-organized through the supervised Fuzzy ARTMAP (FAM) classifier, and the parameters of these nodes are adapted using the Harmonic Means (HM) algorithm. The performance of the proposed model is evaluated empirically using three benchmark data sets. The results indicate that the proposed model is able to produce a compact network structure and, at the same time, to provide high classification performances.

    AB - The radial basis function neural network trained with a dynamic decay adjustment (known as RBFNDDA) algorithm exhibits a greedy insertion behavior as a result of recruiting many hidden nodes for encoding information during its training process. In this chapter, a new variant RBFNDDA is proposed to rectify such deficiency. Specifically, the hidden nodes of RBFNDDA are re-organized through the supervised Fuzzy ARTMAP (FAM) classifier, and the parameters of these nodes are adapted using the Harmonic Means (HM) algorithm. The performance of the proposed model is evaluated empirically using three benchmark data sets. The results indicate that the proposed model is able to produce a compact network structure and, at the same time, to provide high classification performances.

    KW - Adaptive resonance theory

    KW - Classification

    KW - Harmonic mean algorithm

    KW - Radial basis function neural network

    UR - http://www.scopus.com/inward/record.url?scp=84951745921&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84951745921&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-21209-8_4

    DO - 10.1007/978-3-319-21209-8_4

    M3 - Conference contribution

    SN - 9783319212081

    VL - 42

    T3 - Smart Innovation, Systems and Technologies

    SP - 49

    EP - 60

    BT - Smart Innovation, Systems and Technologies

    PB - Springer Science and Business Media Deutschland GmbH

    ER -