• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2005 Fiscal Year Final Research Report Summary

New Approaches for large scale classification problems

Research Project

Project/Area Number 16510106
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeSingle-year Grants
Section一般
Research Field Social systems engineering/Safety system
Research InstitutionTokyo Institute of Technology

Principal Investigator

YAJIMA Yasutoshi  Tokyo Institute of Technology, Graduate School of Decision Science and Technology, Associate Professor, 大学院・社会理工学研究科, 助教授 (80231645)

Project Period (FY) 2004 – 2005
KeywordsClassification / Cutting Planes / Supervised learning / Kernel matrix / Data mining
Research Abstract

We propose an SVM based feature ranking and selection method. In the proposed procedure, several properties of SVMs with RBF kernel functions are exploited to calculate the vectors lying on the discriminate boundary. We show that these vectors, or their gradient vectors, can be calculated efficiently only by the elementary matrix and vector calculation. The results of numerical experiments on the Reuter-21578 dataset show that the proposed method achieves higher classification performance than that based on LSI and $chi^2$ statistics values.
Also, we have introduced semi-supervised learning approaches for partially labeled data points. Our approaches utilize the manifold structure of the given data points, which is characterized as a weighted graph or the associated Laplacian. We show that a number of conventional SVM frameworks such as the 1-norm and 2-norm soft margin formulations and hard margin formulation can be naturally extended to the semi-supervised settings. The resulting formulations are quite simple convex quadratic programming problems. The sparse structure of the graph Laplacian enables us to optimize the problem in a practical amount of computational time even if the number of the variables, i.e., the number of the data points, is very large. Moreover, we show that several existing Laplacian based approaches can be seen as special classes of our framework. The numerical experiments indicate that our approaches perform well on some data sets. Our future plans include experiments on much larger data sets.

  • Research Products

    (10 results)

All 2006 2005 2004

All Journal Article (10 results)

  • [Journal Article] One-Class Support Vector Machines for Recommendation Tasks2006

    • Author(s)
      Yajima, Yasutoshi
    • Journal Title

      LNAI 3918

      Pages: 230-239

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] One-Class Support Vector Machines for Recommendation Tasks2006

    • Author(s)
      Yajima Yasutoshi
    • Journal Title

      LNAI 3918

      Pages: 230-239

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] A Cutting Plane Algorithm for Multiclass Kernel Discriminations2006

    • Author(s)
      Tien-Fang Kuo, Yasutoshi Yajima
    • Journal Title

      GrC2006

      Pages: 223-228

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Ranking Selecting Terms for Text Categorization via SVM Discriminate Boundary2005

    • Author(s)
      Kuo, Tien-Fang, Yajima, Yasutoshi
    • Journal Title

      International Conference on Granular Computing 2005

      Pages: 496-501

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] 百貨店における隠れた親近性の発掘2005

    • Author(s)
      オウロ, 吉原亜弥, 矢島安敬
    • Journal Title

      オペレーションズ・リサーチ 50

      Pages: 164-175

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Optimization Approaches for Semi-Supervised Learning2005

    • Author(s)
      Yajima, Yasutoshi, Hoshida, Takashi
    • Journal Title

      ICMLA 2005

      Pages: 247-252

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Optimization Approaches for Semi-Supervised Learning2005

    • Author(s)
      Yajima, Yasutoshi, Hoshiba, Takashi
    • Journal Title

      ICMLA2005

      Pages: 247-252

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] SVMを使った非線形判別における属性抽出法2004

    • Author(s)
      矢島安敏, 安部哲朗
    • Journal Title

      日本応用数理学会論文誌 14

      Pages: 39-57

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Linear Programming Approaches for Multicategory Support Vector Machines2004

    • Author(s)
      Yajima, Yasutoshi
    • Journal Title

      European Journal of Operational Research 162

      Pages: 514-531

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Linear Programming Approaches for Multicategory Support Vector Machines2004

    • Author(s)
      Yajima, Yasutoshi
    • Journal Title

      EJOR 162

      Pages: 514-531

    • Description
      「研究成果報告書概要(欧文)」より

URL: 

Published: 2007-12-13  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi