DOI: 10.3724/SP.J.1087.2008.00134

Journal of Computer Applications (计算机应用) 2008/28:1 PP.134-135

Utilizing particle swarm optimization to optimize hyper-parameters of SVM classifier

Particle swarm optimization used for optimization selection for hyper-parameter of support vector machine classifier was designed and implemented utilizing global searching property of particle swarm optimization algorithm while the algorithm was used to solve combinatorial optimization problems. The method of individuals coding and evaluating was described in brief. The experimental statistic results demonstrate that the algorithm is effective and efficacious. In the end, some in-depth works are listed on the base of above-mentioned study.

Key words:Support Vector Machine (SVM),classifier,parameter optimization,particle swarm optimization

ReleaseDate:2014-07-21 14:08:19

[1]VAPNIK V. The nature of statistical learning[M]. New York: Springer-Verlag, 1995.

[2]陈果. 基于遗传算法的支持向量机分类器模型参数优化[J]. 机械科学与技术, 2007, 26(3):347-350.

[3]杜京义, 侯媛彬. 基于遗传算法的支持向量回归机参数选取 [J]. 系统工程与电子技术, 2006, 28(9):1430-1433.

[4]周红刚, 杨春德. 基于免疫算法与支持向量机的异常检测方法[J]. 计算机应用, 2006, 26(9):2145-2147.

[5]燕中, 袁春伟. 基于蚁群智能和支持向量机的人脸性别分类方法[J]. 电子与信息学报, 2004, 26(8):1177-1181.

[6]CHANG C C, LIN C J. LIBSVM - A library for support vector machines [CP/OL]., 2001.

[7]CHANG C C, LIN C J. LIBSVM data: Classification, regression, and multi-label [EB/OL]., 2001.