多面体識別子族における識別子選択  [in Japanese] Classifier selection in a family of polyhedron classifiers  [in Japanese]

Search this Article

Author(s)

Abstract

クラス領域の近似において凸包は有効であるものの,高次元で凸包を構成するのは計算量的に困難である.本稿では,次元に線形な計算量で済む近似凸包の構成法を提案する.加えて,近似凸包を含む多面体領域を用いた識別子の族から,与えられた問題に有効な識別子を選択する基準を検討する.

It is an efficient way to approximate a class region by convex hulls of samples. However, the convex hull is computationally hard to be constructed in high dimensions. In this paper, we propose a way to construct an approximate convex hull in a linear time of dimension. In addition, we discuss on selecting an optimal classifier from a family of polyhedron classifiers derived from approximate convex hulls.

Journal

  • IEICE technical report

    IEICE technical report 108(363), 37-41, 2008-12-11

    The Institute of Electronics, Information and Communication Engineers

References:  6

Codes

  • NII Article ID (NAID)
    110007123816
  • NII NACSIS-CAT ID (NCID)
    AN10541106
  • Text Lang
    JPN
  • Article Type
    ART
  • ISSN
    09135685
  • NDL Article ID
    9764258
  • NDL Source Classification
    ZN33(科学技術--電気工学・電気機械工業--電子工学・電気通信)
  • NDL Call No.
    Z16-940
  • Data Source
    CJP  NDL  NII-ELS 
Page Top