Quantum-inspired evolutionary tuning of SVM parameters

作     者：Zhiyong Luo a, Ping Wanga, Yinguo Lia, Wenfeng Zhang b, Wei Tangb, Min Xiang a a School of Automation, Chongqing University of Posts and Telecommunications, Chongqing 400065, China b School of Automation, Northwestern Polytechnical University, Xi’an 710072, China 

作者机构：School of Automation Chongqing University of Posts and Telecommunications Chongqing 400065 China School of Automation Northwestern Polytechnical University Xi’an 710072 China 

出 版 物：《Progress in Natural Science:Materials International》 (自然科学进展·国际材料(英文))

年 卷 期：2008年第18卷第4期

页      面：475-480页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National High-tech Research and Development Program: 2004AA412020 2006AA040302-1 

主　　题：Quantum-inspired evolutionary algorithm (QEA) Parameters tuning Support vector machines (SVM) Least squares support vector machines (LS-SVM) 

摘      要：The most commonly used parameters selection method for support vector machines (SVM) is cross-validation, which needs a long- time complicated calculation. In this paper, a novel regularization parameter and a kernel parameter tuning approach of SVM are pre- sented based on quantum-inspired evolutionary algorithm (QEA). QEA with quantum chromosome and quantum mutation has better global search capacity. The parameters of least squares support vector machines (LS-SVM) can be adjusted using quantum-inspired evo- lutionary optimization. Classification and function estimation are studied using LS-SVM with wavelet kernel and Gaussian kernel. The simulation results show that the proposed approach can effectively tune the parameters of LS-SVM, and the improved LS-SVM with wavelet kernel can provide better precision.