Journal of Computer Applications ›› 2013, Vol. 33 ›› Issue (07): 1894-1897.DOI: 10.11772/j.issn.1001-9081.2013.07.1894

• Network and distributed techno • Previous Articles     Next Articles

Least square support vector classification-regression machine for multi-classification problems

ZHAI Jia,HU Yiqing,XU Er   

  1. School of Mathematics and Physics, University of Science and Technology Beijing, Beijing 100083, China
  • Received:2013-01-30 Revised:2013-02-27 Online:2013-07-06 Published:2013-07-01
  • Contact: HU Yiqing

用于多分类问题的最小二乘支持向量分类—回归机

翟嘉,胡毅庆,徐尔   

  1. 北京科技大学 数理学院,北京 100083
  • 通讯作者: 胡毅庆
  • 作者简介:翟嘉(1989-),女,山西盂县人,硕士研究生,主要研究方向:运筹与控制;胡毅庆(1980-),男,北京人,讲师,博士,主要研究方向:运筹与控制;徐尔(1961-),女,北京人,副教授,主要研究方向:运筹学。
  • 基金资助:

    国家自然科学基金青年基金资助项目(61202184)

Abstract: Tri-class classification method based on Support Vector Machine (SVM) is a kind of method for solving multi-class classification problems. Least Square Support Vector Classification-Regression (LSSVCR) was proposed, which considered the effects of all the sample points by using least squares objective function. Even if there were wrongly marked sample points in the training set, the result would not be affected largely by them. LSSVCR was more accurate and faster, and it was efficient for the problems that there are large differences among the number of sample points in different classes. The numerical experiments show that the proposed method raises the accuracy by 2.57% on average compared to the existing tri-classification methods.

Key words: multi-class classification problem, tri-class classification problem, Least Square Support Vector Machine (LSSVM), classification-regression machine, one versus one versus rest (1-v-1-v-r) method

摘要: 基于支持向量机(SVM)的三分类方法是处理多分类问题的一类方法。提出了最小二乘支持向量分类〖CD*2〗回归机(LSSVCR)算法,通过最小二乘目标函数充分考虑所有样本点对分类的影响,使得训练集中即使有个别样本点被标错类别,对分类结果也不会产生太大的影响,从而提高分类的准确性。该方法能够提高分类的准确率和分类速度,同时算法对于不同类别间样本数目差异较大的情况也有很好的分类效果。数值实验结果表明所提算法是可行的,且与已有的三分类算法相比在分类准确性上平均提高了2.57%,在运算速度上也有了较大的提高。

关键词: 多分类问题, 三分类问题, 最小二乘支持向量机, 分类回归机, 一对一对多方法

CLC Number: