计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (22): 89-92.

• 学术探讨 • 上一篇    下一篇

LS-SVM的组合优化算法研究

曾绍华1,2   

  1. 1.重庆大学 计算机科学学院,重庆 400030
    2.重庆师范大学 经济与管理学院,重庆 400030
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-01 发布日期:2007-08-01
  • 通讯作者: 曾绍华

Research on LS-SVM combination optimization algorithm

ZENG Shao-hua1,2   

  1. 1.College of Computer Science,Chongqing University,Chongqing 400030,China
    2.College of Economy & Management,Chongqing Normal University,Chongqing 400030,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-01 Published:2007-08-01
  • Contact: ZENG Shao-hua

摘要: 根据LS-SVM不具稀疏解的特性,提出了LS-SVM的组合优化算法及其改进算法,并利用整数规划优化了LS-SVM的组合优化算法中划分的样本子集包含的样本个数p。最后,对算法复杂性做了分析,仿真验证了算法的收敛性和有效性。

关键词: LS-SVM, 组合优化算法, 整数规划

Abstract: According as the nature that LS-SVM hasn’t the sparse solution,LS-SVM combination optimization algorithm and its improving is proposed.And the p is optimized with integer programming,which is number of samples in the samples subset based on LS-SVM combination optimization algorithm.Finally,analyzes it’s complexity,and tests and verifies convergence and availability of the algorithms by the simulation.

Key words: LS-SVM, combination optimization algorithm, integer programming