Some Notes on the Parzen Classifier Design

1 Citations 9 References
  • MUTO Yoshihiko
    Ube National College of Technology
  • NAGASE Hirokazu
    Department of Computer Science and Systems Engineering, Faculty of Engineering, Yamaguchi University
  • HAMAMOTO Yoshihiko
    Department of Computer Science and Systems Engineering, Faculty of Engineering, Yamaguchi University

Bibliographic Information

Other Title
  • Parzen 識別器設計における2, 3の考察

Search this article

Description

In this paper, we discuss the effects of the sample size on the generalization ability of Parzen classifiers. When the sizes of samples per class are much unequal, the performance of the Parzen classifier is further degraded. In order to overcome this problem, we propose to use the Toeplitz estimator and bootstrap samples in designing Parzen classifiers. Experimental results show that these techniques are very effective means for designing Parzen classifiers, particularly when the sizes of samples per class are much unequal.

Journal

  • Technical report of IEICE. PRMU

    Technical report of IEICE. PRMU 98 (528), 173-179, 1999-01-22

    The Institute of Electronics, Information and Communication Engineers

Citations (1)*help

See more

References(9)*help

See more

Keywords

Details 詳細情報について

  • CRID
    1571135652438616320
  • NII Article ID
    110003274878
  • NII Book ID
    AN10541106
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top