计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (22): 35-38.DOI: 10.3778/j.issn.1002-8331.2008.22.010

• 博士论坛 • 上一篇    下一篇

基于IGA的支持向量机特征子集选择和参数优化

郝艳友1,2,迟忠先1,李克秋1,张 永3   

  1. 1.大连理工大学 计算机科学与工程系,辽宁 大连 116024
    2.中国建设银行 大连市分行 科技部,辽宁 大连 116011
    3.辽宁师范大学 计算机系,辽宁 大连 116029
  • 收稿日期:2008-03-20 修回日期:2008-04-23 出版日期:2008-07-11 发布日期:2008-07-11
  • 通讯作者: 郝艳友

IGA-based feature subset selection and parameters optimization for support vector machines

HAO Yan-you1,2,CHI Zhong-xian1,LI Ke-qiu1,ZHANG Yong3   

  1. 1.Department of Computer Science and Engineering,Dalian University of Technology,Dalian,Liaoning 116024,China
    2.Dalian Branch,China Construction Bank,Dalian,Liaoning 116011,China
    3.Department of Computer,Liaoning Normal University,Dalian,Liaoning 116029,China
  • Received:2008-03-20 Revised:2008-04-23 Online:2008-07-11 Published:2008-07-11
  • Contact: HAO Yan-you

摘要: 特征子集选择和训练参数的优化一直是SVM研究中的两个重要方面,选择合适的特征和合理的训练参数可以提高SVM分类器的性能,以往的研究是将两个问题分别进行解决。随着遗传优化等自然计算技术在人工智能领域的应用,开始出现特征选择及参数的同时优化研究。研究采用免疫遗传算法(IGA)对特征选择及SVM 参数的同时优化,提出了一种IGA-SVM 算法。实验表明,该方法可找出合适的特征子集及SVM 参数,并取得较好的分类效果,证明算法的有效性。

关键词: 支持向量机, 特征选择, 参数优化, 免疫遗传算法

Abstract: Feature selection and parameter optimization are two important aspects for improving classifier performance.However,they are solved separately traditionally.Recently,with the wide applications of evolutionary computation in pattern recognition area,simultaneous feature selection and parameter optimization become possible and tendency.To solve this problem,a simultaneous feature selection and SVM parameter optimization algorithm based on Immune GA algorithm is proposed,named as IGA-SVM.The experimental results show that the algorithm can efficiently find the suitable feature subsets and SVM parameters,which result in a good classification performance.

Key words: Support Vector Machine(SVM), feature selection, parameters optimization, Immune Genetic Algorithm(IGA)