Function and surface approximation based on enhanced Kernel Regression for small sample sets

Mohd Ibrahim Shapiai, Zuwairie Ibrahim, Marzuki Khalid, Lee Wen Jau, Vladimir Pavlovic, Junzo Watada

    Research output: Contribution to journalArticle

    7 Citations (Scopus)

    Abstract

    The function approximation problem is to find the appropriate relationship between a dependent and independent variable(s). Function approximation algorithms generally require sufficient samples to approximate a function. Insufficient samples may cause any function approximation algorithm to result in unsatisfactory predictions. To solve this problem, a function approximation algorithm called Weighted Kernel Regression (WKR), which is based on Nadaraya-Watson kernel regression (NWKR), is proposed. In the proposed framework, the original NWKR algorithm is enhanced by expressing the observed samples in a square kernel matrix. The WKR is trained to estimate the weight for the testing phase. The weight is estimated iteratively and governed by the error function to find a good approximation model. Four experiments are conducted to show the capability of the WKR. The results show that the proposed WKR model is effective in cases where the target function is non-linear and the given training sample is small. The performance of the WKR is also compared with other existing function approximation algorithms, such as artificial neural networks (ANN).

    Original languageEnglish
    Pages (from-to)5947-5960
    Number of pages14
    JournalInternational Journal of Innovative Computing, Information and Control
    Volume7
    Issue number10
    Publication statusPublished - 2011 Oct

    Fingerprint

    Surface Approximation
    Kernel Regression
    Function Approximation
    Small Sample
    Approximation algorithms
    Approximation Algorithms
    Error function
    Approximation Problem
    Training Samples
    Artificial Neural Network
    Regression Model
    Sufficient
    kernel
    Testing
    Target
    Neural networks
    Dependent
    Prediction
    Approximation

    Keywords

    • Artificial neural network
    • Non-linear function
    • Small samples
    • Weighted kernel regression

    ASJC Scopus subject areas

    • Computational Theory and Mathematics
    • Information Systems
    • Software
    • Theoretical Computer Science

    Cite this

    Shapiai, M. I., Ibrahim, Z., Khalid, M., Jau, L. W., Pavlovic, V., & Watada, J. (2011). Function and surface approximation based on enhanced Kernel Regression for small sample sets. International Journal of Innovative Computing, Information and Control, 7(10), 5947-5960.

    Function and surface approximation based on enhanced Kernel Regression for small sample sets. / Shapiai, Mohd Ibrahim; Ibrahim, Zuwairie; Khalid, Marzuki; Jau, Lee Wen; Pavlovic, Vladimir; Watada, Junzo.

    In: International Journal of Innovative Computing, Information and Control, Vol. 7, No. 10, 10.2011, p. 5947-5960.

    Research output: Contribution to journalArticle

    Shapiai, MI, Ibrahim, Z, Khalid, M, Jau, LW, Pavlovic, V & Watada, J 2011, 'Function and surface approximation based on enhanced Kernel Regression for small sample sets', International Journal of Innovative Computing, Information and Control, vol. 7, no. 10, pp. 5947-5960.
    Shapiai, Mohd Ibrahim ; Ibrahim, Zuwairie ; Khalid, Marzuki ; Jau, Lee Wen ; Pavlovic, Vladimir ; Watada, Junzo. / Function and surface approximation based on enhanced Kernel Regression for small sample sets. In: International Journal of Innovative Computing, Information and Control. 2011 ; Vol. 7, No. 10. pp. 5947-5960.
    @article{412a496b669c4dcd96137f5cf1169ca6,
    title = "Function and surface approximation based on enhanced Kernel Regression for small sample sets",
    abstract = "The function approximation problem is to find the appropriate relationship between a dependent and independent variable(s). Function approximation algorithms generally require sufficient samples to approximate a function. Insufficient samples may cause any function approximation algorithm to result in unsatisfactory predictions. To solve this problem, a function approximation algorithm called Weighted Kernel Regression (WKR), which is based on Nadaraya-Watson kernel regression (NWKR), is proposed. In the proposed framework, the original NWKR algorithm is enhanced by expressing the observed samples in a square kernel matrix. The WKR is trained to estimate the weight for the testing phase. The weight is estimated iteratively and governed by the error function to find a good approximation model. Four experiments are conducted to show the capability of the WKR. The results show that the proposed WKR model is effective in cases where the target function is non-linear and the given training sample is small. The performance of the WKR is also compared with other existing function approximation algorithms, such as artificial neural networks (ANN).",
    keywords = "Artificial neural network, Non-linear function, Small samples, Weighted kernel regression",
    author = "Shapiai, {Mohd Ibrahim} and Zuwairie Ibrahim and Marzuki Khalid and Jau, {Lee Wen} and Vladimir Pavlovic and Junzo Watada",
    year = "2011",
    month = "10",
    language = "English",
    volume = "7",
    pages = "5947--5960",
    journal = "International Journal of Innovative Computing, Information and Control",
    issn = "1349-4198",
    publisher = "IJICIC Editorial Office",
    number = "10",

    }

    TY - JOUR

    T1 - Function and surface approximation based on enhanced Kernel Regression for small sample sets

    AU - Shapiai, Mohd Ibrahim

    AU - Ibrahim, Zuwairie

    AU - Khalid, Marzuki

    AU - Jau, Lee Wen

    AU - Pavlovic, Vladimir

    AU - Watada, Junzo

    PY - 2011/10

    Y1 - 2011/10

    N2 - The function approximation problem is to find the appropriate relationship between a dependent and independent variable(s). Function approximation algorithms generally require sufficient samples to approximate a function. Insufficient samples may cause any function approximation algorithm to result in unsatisfactory predictions. To solve this problem, a function approximation algorithm called Weighted Kernel Regression (WKR), which is based on Nadaraya-Watson kernel regression (NWKR), is proposed. In the proposed framework, the original NWKR algorithm is enhanced by expressing the observed samples in a square kernel matrix. The WKR is trained to estimate the weight for the testing phase. The weight is estimated iteratively and governed by the error function to find a good approximation model. Four experiments are conducted to show the capability of the WKR. The results show that the proposed WKR model is effective in cases where the target function is non-linear and the given training sample is small. The performance of the WKR is also compared with other existing function approximation algorithms, such as artificial neural networks (ANN).

    AB - The function approximation problem is to find the appropriate relationship between a dependent and independent variable(s). Function approximation algorithms generally require sufficient samples to approximate a function. Insufficient samples may cause any function approximation algorithm to result in unsatisfactory predictions. To solve this problem, a function approximation algorithm called Weighted Kernel Regression (WKR), which is based on Nadaraya-Watson kernel regression (NWKR), is proposed. In the proposed framework, the original NWKR algorithm is enhanced by expressing the observed samples in a square kernel matrix. The WKR is trained to estimate the weight for the testing phase. The weight is estimated iteratively and governed by the error function to find a good approximation model. Four experiments are conducted to show the capability of the WKR. The results show that the proposed WKR model is effective in cases where the target function is non-linear and the given training sample is small. The performance of the WKR is also compared with other existing function approximation algorithms, such as artificial neural networks (ANN).

    KW - Artificial neural network

    KW - Non-linear function

    KW - Small samples

    KW - Weighted kernel regression

    UR - http://www.scopus.com/inward/record.url?scp=80053555784&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=80053555784&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:80053555784

    VL - 7

    SP - 5947

    EP - 5960

    JO - International Journal of Innovative Computing, Information and Control

    JF - International Journal of Innovative Computing, Information and Control

    SN - 1349-4198

    IS - 10

    ER -