计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (17): 167-169.DOI: 10.3778/j.issn.1002-8331.2009.17.050

• 图形、图像、模式识别 • 上一篇    下一篇

非平衡二叉树多类支持向量机分类方法

夏思宇,潘 泓,金立左   

  1. 东南大学 自动化学院 南京 210096
  • 收稿日期:2008-04-14 修回日期:2008-07-07 出版日期:2009-06-11 发布日期:2009-06-11
  • 通讯作者: 夏思宇

Multi-class SVM method based on non-balanced binary tree

XIA Si-yu,PAN Hong,JIN Li-zuo   

  1. School of Automation,Southeast University,Nanjing 210096,China
  • Received:2008-04-14 Revised:2008-07-07 Online:2009-06-11 Published:2009-06-11
  • Contact: XIA Si-yu

摘要: 提出一种新的基于非平衡二叉树的支持向量机多类别分类方法。该方法通过分析已知类别样本的先验分布知识,构造一个二叉决策树,使容易区分的类别从根节点开始逐层分割出来,以获得较高的推广能力。该方法解决了传统分类算法中所存在的不可分区域问题,在训练时只需构造N-1个SVM分类器,而测试时的判决次数小于N。将该方法应用于人脸识别实验。测试结果表明,与传统分类算法相比,该方法的平均分类时间是最少的。

Abstract: In this paper,a non-balanced binary tree is proposed for extending support vector machines(SVM) to multi-class problems.The non-balanced binary tree is constructed based on the prior distribution of samples,which can make the more separable classes separated at the upper node of the binary tree.For an N class problem,this method only needs N-1 SVM classifiers in the training phase,while it has less than N binary test when making a decision.Further,this method can avoid the unclassifiable regions that exist in the conventional SVMs.The experimental result indicates that maintaining comparable accuracy,this method is faster than other methods in classification.