• DocumentCode
    2095110
  • Title

    A Modified Self-Training Semi-supervised SVM Algorithm

  • Author

    Jin, Yun ; Ma, Yong ; Zhao, Li

  • Author_Institution
    Sch. of Phys. & Electron. Eng., Xuzhou Normal Univ., Xuzhou, China
  • fYear
    2012
  • fDate
    11-13 May 2012
  • Firstpage
    224
  • Lastpage
    228
  • Abstract
    In this paper, we present a modified self-training semi-supervised SVM algorithm. In order to demonstrate its validity and effectiveness, we carry out some experiments which prove that our method is better than the former algorithm. Using our modified self-training semi-supervised SVM algorithm, we can save much time for labeling the unlabelled data.
  • Keywords
    data handling; learning (artificial intelligence); pattern classification; support vector machines; SVM algorithm; modified self-training semi-supervised learning; support vector machine; unlabelled data; Classification algorithms; Convergence; Data models; Iris recognition; Optimization; Support vector machines; Training; SVM; UCI; self-training; semi-supervised learning;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communication Systems and Network Technologies (CSNT), 2012 International Conference on
  • Conference_Location
    Rajkot
  • Print_ISBN
    978-1-4673-1538-8
  • Type

    conf

  • DOI
    10.1109/CSNT.2012.56
  • Filename
    6200629