摘要
支持向量机的性能与核函数的参数及惩罚系数C有很大关系.利用Lozi’s映射的较好遍历性,在粒子群优化算法中引入Lozi’s映射的混沌思想,提出基于混沌粒子群优化算法的SVM参数优化方法.仿真实验表明,该算法能有效提高整个迭代搜索的收敛速度和精度,从而更好地优化SVM参数.
The properties of support vector machine(SVM) have very much relationship of kernel function the parameters and punish coefficient C.Because the Lozi's has good ergodicity of the mapping,this paper introduced the Lozi's mapping based on particle swarm optimization algorithm,and put forward the idea of SVM parameters optimization method based on chaotic particle swarm optimization algorithm.Simulation experiments show that the algorithm can effectively improve the convergence speed and accuracy in whole iteration search,so as to optimize SVM parameters very well.
出处
《重庆文理学院学报(自然科学版)》
2011年第4期81-84,共4页
Journal of Chongqing University of Arts and Sciences
关键词
支持向量机
混沌粒子群
参数优化
support vector machine
chaotic particle swarm
parameter optimization