FEATURE SELECTION WITH NEURAL NETWORKS

• Leray Philippe

  LIP6-Pôle IA-Université

• Gallinari Patrick

  LIP6-Pôle IA-Université

抄録

The observed features of a given phenomenon are not all equally informative : some may be noisy, others correlated or irrelevant. The purpose of feature selection is to select a set of features pertinent to a given task. This is a complex process, but it is an important issue in many fields. In neural networks, feature selection has been studied for the last ten years, using conventional and original methods. This paper is a review of neural network approaches to feature selection. We first briefly introduce baseline statistical methods used in regression and classification. We then describe families of methods which have been developed specifically for neural networks. Representative methods are then compared on different test problems.

収録刊行物

• Behaviormetrika

  Behaviormetrika 26 (1), 145-166, 1999

  日本行動計量学会

キーワード

• Feature Selection
• Subset selection
• Variable Sensitivity
• Sequential Search

詳細情報

• CRID: 1390001205112121088
• NII論文ID: 130003527195
• DOI: 10.2333/bhmk.26.145
• ISSN: 13496964
• ISSN: 03857417
• Web Site: http://link.springer.com/content/pdf/10.2333/bhmk.26.145.pdf
• Web Site: http://link.springer.com/article/10.2333/bhmk.26.145/fulltext.html
• 本文言語コード: en
• データソース種別:
  • JaLC
  • Crossref
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可