作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (19): 181-183. doi: 10.3969/j.issn.1000-3428.2009.19.060

• 人工智能及识别技术 • 上一篇    下一篇

模糊支持向量机的偏移量计算方法

陈家德,吴小俊   

  1. (江南大学信息工程学院,无锡 214122)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-10-05 发布日期:2009-10-05

Offset Calculation Method for Fuzzy Support Vector Machine

CHEN Jia-de, WU Xiao-jun   

  1. (School of Information Technology, Jiangnan University, Wuxi 214122)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-10-05 Published:2009-10-05

摘要: 偏移量确定了支持向量机和模糊支持向量机(FSVM)的最优分类面位置,对分类性能具有较大影响。为提高模糊支持向量机的识别率,基于Fisher判别分析方法提出一种新的偏移量计算方法,将其用于FSVM多类分类器设计。对3种数据集的测试结果表明,使用新偏移量的FSVM识别率高于使用标准偏移量的FSVM识别率。

关键词: 偏移量, 支持向量机, 模糊支持向量机, 机器学习

Abstract: Offset determines the position of optimal separating plane of Support Vector Machine(SVM) and Fuzzy Support Vector Machine(FSVM) and affects the performance of classification greatly. In order to improve the recognition rate of FSVM, this paper proposes a new calculation approach for offset based on Fisher discriminant analysis method and uses it to design FSVM multi-classification. Test results of three data sets show that the recognition rate of FSVM using new offset is higher than the one using normal offset.

Key words: offset, Support Vector Machine(SVM), Fuzzy Support Vector Machine(FSVM), machine learning

中图分类号: