Midpoint-Validation Method for Support Vector Machine Classification

• TAMURA Hiroki

  Faculty of Engineering, University of Miyazaki The Institute of Electronics, Information and Communication Engineers

• TANNO Koichi

  Faculty of Engineering, University of Miyazaki The Institute of Electronics, Information and Communication Engineers

抄録

In this paper, we propose a midpoint-validation method which improves the generalization of Support Vector Machine. The proposed method creates midpoint data, as well as a turning adjustment parameter of Support Vector Machine using midpoint data and previous training data. We compare its performance with the original Support Vector Machine, Multilayer Perceptron, Radial Basis Function Neural Network and also tested our proposed method on several benchmark problems. The results obtained from the simulation shows the effectiveness of the proposed method.

収録刊行物

• IEICE Transactions on Information and Systems

  IEICE Transactions on Information and Systems E91-D (7), 2095-2098, 2008

  一般社団法人 電子情報通信学会

キーワード

• support vector machine
• midpoint-validation
• pattern classification problem

詳細情報

• CRID: 1390001204377760768
• NII論文ID: 10026805215
• NII書誌ID: AA10826272
• DOI: 10.1093/ietisy/e91-d.7.2095
• ISSN: 17451361
• ISSN: 09168532
• HANDLE: 10458/3268
• Web Site: https://miyazaki-u.repo.nii.ac.jp/records/981
• Web Site: https://www.jstage.jst.go.jp/article/transinf/E91.D/7/E91.D_7_2095/_pdf
• 本文言語コード: en
• データソース種別:
  • JaLC
  • IRDB
  • Crossref
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可