Tuning Fuzzy Systems by Function Approximation

D.JA. Bijwaard, M. Poel, N. Ulder

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    Abstract

    A prototype fuzzy system is quite easy to set up and modify with the techniques within fuzzy system theory when linguistic rules can be given for the desired behavior. Fine-tuning of such a system proves more difficult as well as building a system when no rules can be given. Because this tuning is a very specialistic job, during construction as well as in maintenance, ways were examined to do the tuning automatically. Function approximation is described as a general technique to tune parts of a fuzzy systems based on the desired input and output behavior. Two distinct function approximation techniques were taken into consideration: Techniques based on B-splines (analytically as well as numerically) and techniques based on sigmoids (only numerically, e.g. with the backpropagation procedure used in neural networks). Experiments were done to determine which technique gives the best results when inhomogeneously scattered data points with noise are used. The sigmoid technique seemed give the best approximation in areas were no data points were available (generalization). Some constraints could be given for the position and number of B-splines with respect to the distribution and the amount of the data respectively.
    Original languageEnglish
    Title of host publicationProceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996)
    Place of PublicationMarseille, France
    Pages231-237
    Publication statusPublished - 18 Jan 1996
    Event4th International Conference on Neural Networks and their Applications, NEURAP 1996 - Marseilles, France
    Duration: 20 Mar 199622 Mar 1996
    Conference number: 4

    Conference

    Conference4th International Conference on Neural Networks and their Applications, NEURAP 1996
    Abbreviated titleNEURAP
    CountryFrance
    CityMarseilles
    Period20/03/9622/03/96

    Fingerprint

    Fuzzy systems
    Tuning
    Splines
    System theory
    Backpropagation
    Linguistics
    Neural networks
    Experiments

    Keywords

    • METIS-119210

    Cite this

    Bijwaard, D. JA., Poel, M., & Ulder, N. (1996). Tuning Fuzzy Systems by Function Approximation. In Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996) (pp. 231-237). Marseille, France.
    Bijwaard, D.JA. ; Poel, M. ; Ulder, N. / Tuning Fuzzy Systems by Function Approximation. Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996). Marseille, France, 1996. pp. 231-237
    @inproceedings{a69b1b8d20c44acda9e2c3904c2cac57,
    title = "Tuning Fuzzy Systems by Function Approximation",
    abstract = "A prototype fuzzy system is quite easy to set up and modify with the techniques within fuzzy system theory when linguistic rules can be given for the desired behavior. Fine-tuning of such a system proves more difficult as well as building a system when no rules can be given. Because this tuning is a very specialistic job, during construction as well as in maintenance, ways were examined to do the tuning automatically. Function approximation is described as a general technique to tune parts of a fuzzy systems based on the desired input and output behavior. Two distinct function approximation techniques were taken into consideration: Techniques based on B-splines (analytically as well as numerically) and techniques based on sigmoids (only numerically, e.g. with the backpropagation procedure used in neural networks). Experiments were done to determine which technique gives the best results when inhomogeneously scattered data points with noise are used. The sigmoid technique seemed give the best approximation in areas were no data points were available (generalization). Some constraints could be given for the position and number of B-splines with respect to the distribution and the amount of the data respectively.",
    keywords = "METIS-119210",
    author = "D.JA. Bijwaard and M. Poel and N. Ulder",
    year = "1996",
    month = "1",
    day = "18",
    language = "English",
    pages = "231--237",
    booktitle = "Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996)",

    }

    Bijwaard, DJA, Poel, M & Ulder, N 1996, Tuning Fuzzy Systems by Function Approximation. in Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996). Marseille, France, pp. 231-237, 4th International Conference on Neural Networks and their Applications, NEURAP 1996, Marseilles, France, 20/03/96.

    Tuning Fuzzy Systems by Function Approximation. / Bijwaard, D.JA.; Poel, M.; Ulder, N.

    Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996). Marseille, France, 1996. p. 231-237.

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Tuning Fuzzy Systems by Function Approximation

    AU - Bijwaard, D.JA.

    AU - Poel, M.

    AU - Ulder, N.

    PY - 1996/1/18

    Y1 - 1996/1/18

    N2 - A prototype fuzzy system is quite easy to set up and modify with the techniques within fuzzy system theory when linguistic rules can be given for the desired behavior. Fine-tuning of such a system proves more difficult as well as building a system when no rules can be given. Because this tuning is a very specialistic job, during construction as well as in maintenance, ways were examined to do the tuning automatically. Function approximation is described as a general technique to tune parts of a fuzzy systems based on the desired input and output behavior. Two distinct function approximation techniques were taken into consideration: Techniques based on B-splines (analytically as well as numerically) and techniques based on sigmoids (only numerically, e.g. with the backpropagation procedure used in neural networks). Experiments were done to determine which technique gives the best results when inhomogeneously scattered data points with noise are used. The sigmoid technique seemed give the best approximation in areas were no data points were available (generalization). Some constraints could be given for the position and number of B-splines with respect to the distribution and the amount of the data respectively.

    AB - A prototype fuzzy system is quite easy to set up and modify with the techniques within fuzzy system theory when linguistic rules can be given for the desired behavior. Fine-tuning of such a system proves more difficult as well as building a system when no rules can be given. Because this tuning is a very specialistic job, during construction as well as in maintenance, ways were examined to do the tuning automatically. Function approximation is described as a general technique to tune parts of a fuzzy systems based on the desired input and output behavior. Two distinct function approximation techniques were taken into consideration: Techniques based on B-splines (analytically as well as numerically) and techniques based on sigmoids (only numerically, e.g. with the backpropagation procedure used in neural networks). Experiments were done to determine which technique gives the best results when inhomogeneously scattered data points with noise are used. The sigmoid technique seemed give the best approximation in areas were no data points were available (generalization). Some constraints could be given for the position and number of B-splines with respect to the distribution and the amount of the data respectively.

    KW - METIS-119210

    M3 - Conference contribution

    SP - 231

    EP - 237

    BT - Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996)

    CY - Marseille, France

    ER -

    Bijwaard DJA, Poel M, Ulder N. Tuning Fuzzy Systems by Function Approximation. In Proceedings of the 4th International Conference on Neural Networks and their Applications (NEURAP 1996). Marseille, France. 1996. p. 231-237