期刊文献+

基于最小二乘的孪生有界支持向量机分类算法 被引量:8

Least squares based twin bounded support vector machine classification algorithm
原文传递
导出
摘要 基于经典的孪生有界支持向量机(TBSVM),构造了一个既简单又快速的基于最小二乘的孪生有界支持向量机(LSTBSVM)的二分类算法.该算法简单地将TBSVM模型中的两个目标函数中不等式约束问题修改为等式约束,问题最终归结为求解两个最小二乘问题,以至于两个最优不平行平面可通过求解一对线性方程组获取.与TBSVM相比,LSTBSVM具有更低的时间复杂度,以至于可以有效地处理大数据集.通过理论分析和在传统的UCI和人工数据集上的实验显示,LSTBSVM不仅具较快的计算速度,且能得到与TBSVM相当的性能. Based on the classic twin bounded support vector machine(TBSVM),an efficient but simple least squares twin bounded support vector machines(LSTBSVM) for binary classification problem was proposed.This algorithm reformulated the inequality constraints from the TBSVM problems by using equality constraints,reducing to solve a pair of least squares problems.Finally,only a pair of systems of linear equations is needed to be solved for the derivation of two optimal nonparallel planes.Compared to TBSVM,LSTBSVM has low time complexity such that can effectively handle larger datasets.Reliable theoretical analysis and extensive experiments on UCI and artificial datasets show that LSTBSVM is fast computationally and obtain the comparable classification performance to TBSVM.
作者 业巧林 闫贺 Ye Qiaolin ,Yan He(College of Information Science and Technology, Nanjing Forestry University, Nanjing 210037, Chin)
出处 《华中科技大学学报(自然科学版)》 EI CAS CSCD 北大核心 2018年第3期30-35,共6页 Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金 国家自然科学基金资助项目(61401214) 江苏省自然科学基金资助项目(BK20171453)
关键词 支持向量机 模式分类 最小二乘问题 孪生有界支持向量机 大规模数据 非平行平面 support vector machines pattern classification least squares problems twin bounde support vector machineslarge-scale data nonparallel planes
  • 相关文献

参考文献4

二级参考文献41

  • 1杨绪兵,陈松灿.基于原型超平面的多类最接近支持向量机[J].计算机研究与发展,2006,43(10):1700-1705. 被引量:16
  • 2Fung G, Mangasarian O L. Proximal support vector machine classifiers [C] //Proc of Knowledge Discovery and Data Mining(KDD2001). New York: ACM, 2001:77-86.
  • 3Mangasarian O L, Wild E. Multisurface proximal support vector machine classification via generalized eigenvalues [J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2006, 28(1): 69-74.
  • 4Jayadeva, Khemchandai R, Chandra S. Fuzzy proximal support vector classification via generalized cigenvalues [G]// LNCS 3776: Berlin: Springer, 2005:360-363.
  • 5Guarracino M R, Cifarelli C, Seref O, et al. A classification method based on generalized eigenvalue problems [J]. Optimization Methods and Software, 2007, 22(1): 73-81.
  • 6Jayadeva, Khernchandai R, Chandra S. Twin support vector machines for pattern classification [J]. IEEE Trans on Pattern Analysis and Machine Intelligence (TPAMI), 2007, 29(5) : 905-910.
  • 7Zou H. An improved 1-norm SVM for simultaneous classification and variable selection [J]. Journal of Machine Learning Research Proceedings Track, 2007, 2:675-681.
  • 8Sun L, Ji S W, Ye J P. A Least Squares Formation for a Class of Generalized Eigenvalue Problems in Machine Learning [C/OL]//Proc of the 26th Int Conf on Machine Learning. 2009 [2009-11-01]. http://www. machinelearning. org/arehive/ieml2009/.
  • 9Fiacco A V, McCormick G P. Nonlinear Programming: Sequential Unconstrained Minimization Techniques [M]. New York: John Wiley & Sons, 1968.
  • 10Mangasarian O L. A finite Newton method for classification problems [J]. Optimization Methods and Software, 2002, 17:913-929.

共引文献51

同被引文献58

引证文献8

二级引证文献31

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