본문 바로가기
  • Home

A Study on automatic assignment of descriptors using machine learning

  • Journal of the Korean Society for Information Management
  • Abbr : JKOSIM
  • 2006, 23(1), pp.279~299
  • DOI : 10.3743/KOSIM.2006.23.1.279
  • Publisher : 한국정보관리학회
  • Research Area : Interdisciplinary Studies > Library and Information Science
  • Received : February 27, 2006
  • Accepted : March 16, 2006
  • Published : March 30, 2006

Kim, Pan Jun 1

1신라대학교

Accredited

ABSTRACT

This study utilizes various approaches of machine learning in the process of automatically assigning descriptors to journal articles. After selecting core journals in the field of information science and organizing test collection from the articles of the past 11 years, the effectiveness of feature selection and the size of training set was examined. In the regard of feature selection, after reducing the feature set by χ2 statistics(CHI) and criteria which prefer high-frequency features(COS, GSS, JAC), the trained Support Vector Machines(SVM) performs the best. With respective to the size of the training set, it significantly influences the performance of Support Vector Machines(SVM) and Voted Perceptron(VTP). but it scarcely affects that of Naive Bayes(NB).

Citation status

* References for papers published after 2023 are currently being built.