计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (8): 12-15.

• 博士论坛 • 上一篇    下一篇

一类直接构造的模糊多类支持向量分类器

张 永1,2,迟忠先2,米 滢3   

  1. 1.辽宁师范大学 计算机系,辽宁 大连 116029
    2.大连理工大学 计算机科学与工程系,辽宁 大连 116024
    3.大连职业技术学院 基础部,辽宁 大连 116035
  • 收稿日期:2007-10-28 修回日期:2007-12-05 出版日期:2008-03-11 发布日期:2008-03-11
  • 通讯作者: 张 永

Novel direct constructing-based fuzzy multi-class support vector classifier

ZHANG Yong1,2,CHI Zhong-xian2,MI Ying3   

  1. 1.Department of Computer,Liaoning Normal University,Dalian,Liaoning 116029,China
    2.Department of Computer Science and Engineering,Dalian University of Technology,Dalian,Liaoning 116024,China
    3.Foundation Department,Dalian Vocational Technical College,Dalian,Liaoning 116035,China
  • Received:2007-10-28 Revised:2007-12-05 Online:2008-03-11 Published:2008-03-11
  • Contact: ZHANG Yong

摘要: 针对直接多类分类方法,提出了一种新的基于直接构造多类SVM分类器的模糊多类支持向量机算法FCS-SVM。在算法中,重构了优化问题及其约束条件,以及Lagrange公式,并进行了推导。通过在标准数据集上的几个实验,对这些算法进行了比较分析。实验结果表明提出的算法可以得到比较理想的分类精度。

Abstract: This paper presents a novel direct constructing-based fuzzy multi-class support vector classifier,named as FCS-SVM,based on previous multi-class classification method by Crammer and Singer.This algorithm,the optimal problem and its constraints of multi-class classification are reconstructed,and its corresponding Lagrangian formula is re-deduced.Experimental comparison with the previous work indicates that the method can obtain better classification ratio.