ファジィクラスタリングと多変量解析  [in Japanese] FUZZY CLUSTERING AND MULTIVARIATE ANALYSIS  [in Japanese]

Access this Article

Search this Article

Author(s)

    • 佐藤 美佳 Sato Mika
    • 筑波大学大学院システム情報工学研究科リスク工学専攻 Faculty of Systems and Information Engineering, University of Tsukuba

Abstract

従来より, ファジィ理論を既存の多変量解析に適用し, 現実のデータに内在する不確実性, あるいは, 方法論そのものの不確実性を積極的に取り入れようとするファジィ多変量解析の方法が提案されている.本稿では, ファジィ多変量解析の中で, 特にファジィクラスタリングについて概説し, 近年, 筆者らにより提案されているファジィクラスタリングと既存の多変量解析手法とのハイブリッド手法について紹介する.

Conventionally, fuzzy multivariate data analysis has been proposed along with the issue of positively introducing uncertainty in real data and the methodology itself. Fuzzy Clustering is one method which can capture the uncertainty situation of real data and it is well known that fuzzy clustering can obtain a robust result as compared with conventional hard clustering. Following along with the emphasis on the general problem of data analysis, which is a solution for analyzing a huge amount of complex data, the merit of fuzzy clustering for this is expected. In this paper, we describe fuzzy clustering methods, which are methods in fuzzy multivariate analysis, along with several hybrid methods of fuzzy clustering and conventional multivariate analysis which have recently been proposed by us based on the idea that the multiple merits of methods can cope with the inherent classification structures.

Journal

  • Bulletin of the Computational Statistics of Japan

    Bulletin of the Computational Statistics of Japan 17(2), 147-156, 2005

    Japanese Society of Computational Statistics

References:  31

Cited by:  1

Codes

  • NII Article ID (NAID)
    110002325568
  • NII NACSIS-CAT ID (NCID)
    AN10195854
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    0914-8930
  • NDL Article ID
    7718596
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-1382
  • Data Source
    CJP  CJPref  NDL  NII-ELS  J-STAGE 
Page Top