期刊文献+

基于双支持向量机的偏二叉树多类分类算法 被引量:28

A partial binary tree algorithm for multiclass classification based on twin support vector machines
下载PDF
导出
摘要 提出一种基于双支持向量机的偏二叉树多类分类算法,偏二叉树双支持向量机多类分类算法.该算法综合了二叉树支持向量机和双支持向量机的优势,实现了在不降低分类性能的前提下,大大缩短训练时间.理论分析和UCI(University of California Irvine)机器学习数据库数据集上的实验结果共同证明,偏二叉树双支持向量机多类分类算法在训练时间上具有绝对的优势,尤其在处理稍大数据集的多类分类问题时,这一优势尤为突出;实验仿真结果还证明,在采用非线性核时,该算法取得了比基于经典支持向量机的一对其余多类分类算法及二叉树支持向量机更好的分类效果;同时该算法还解决了后两种算法可能存在的样本不平衡问题,以及基于经典支持向量机的一对其余多类分类算法可能存在的不可分区域问题. A new algorithm for multiclass classification problem is presented in this paper.This algorithm,referred here as PBT-TSVM(partial binary tree and twin support vector machines),is a combination of the advantages of binary tree support vector machines(BT-SVM) with those of twin support vector machines(TSVM).Theoretical analysis and experimental results on UCI datasets prove that our PBT-TSVM algorithm not only significantly reduces the training time especially on large datasets,but also gets better classification accuracy via nonlinear kernel functions than 1-v-r SVM(one-versus-rest support vector machines) and BT-SVM.At the same time,our algorithm solves the potential problem of unbalanced dataset of 1-v-r SVM and BT-SVM when dealing with multiclass classifications.Furthermore,it solves the potential problem existing in 1-v-r SVM that some samples cannot be classified.
出处 《南京大学学报(自然科学版)》 CAS CSCD 北大核心 2011年第4期354-363,共10页 Journal of Nanjing University(Natural Science)
基金 中央高校基本科研业务费专项资金(GK200901006) 中央高校基本科研业务费专项资金(GK201001003) 陕西省自然科学基础研究计划(2010JM3004)
关键词 双支持向量机 偏二叉树支持向量机 支持向量机 多类分类 multiclass classification binary tree support vector machines twin support vector machines support vector machines
  • 相关文献

参考文献25

  • 1Vapnik V N. The nature of statistical learning theory. New York:Springer, 2000.
  • 2Burges C J C. A tutorial on support vector ma- chines for pattern recognition. Data Mining and knowledge Discovery, 1998, 2(2): 121-167.
  • 3Mangasarian O L, Musicant D R. Lagrangian support vector machines. Journal of Machine Learning Research, 2001, 1:161-177.
  • 4Lee Y J, Mangasarian O L. SSVM: A smooth support vector machine for classification. Com- putational Optimization and Applications, 2001, 20(1) : 5-22.
  • 5Lee Y J, Mangasarian O L. RSVM: Reduced support vector machines. Proceedings of First SIAM International Conference on Data Mining,2001: 1-17.
  • 6Suykens J A K, Vandewalle J. Least squares support vector machine classifier. Neural pro- cessing Letters, 1999, 9(3): 293-300.
  • 7Mangasarian O L, Wild E W. Multisurface proximal support vector classification via gener- alized eigenvalues. IEEE transactions on Pattern Analysis and Machine Intelligence, 2006, 28 (1) : 69-74.
  • 8Fun G, Mangasarian O L. Proximal support vector machine classifiers. Proceedings KDD 2001.. Knowledge Discovery and Data Mining, 2001: 77-86.
  • 9Jayadeva, Khemchandani R, Chandra S. Twin support vector machines for pattern classifica- tion. IEEE transactions on Pattern Analysis and Machine Intelligence, 2007, 29(5):905-910.
  • 10Cristianini N, Shawe-Taylor J. An introduction to support vector machines and other kernel- based learning methods. Beijing: China Machine Press, 2005.

同被引文献213

引证文献28

二级引证文献122

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部