基于混沌粒子群优化的支持向量机训练方法

为使粒子群优化算法初始粒子均匀分布在解空间，通过对混沌运动的遍历性和粒子群优化算法中惯性权重的分析，提出一种混沌粒子群算法。该算法对Circle模型进行改进，将其引入粒子群算法中，避免了粒子群算法陷入局部最优。给出应用混沌粒子群算法训练SVM的方法，并将其应用于人脸识别。仿真实验结果表明，改进的CPSO SVM方法比CPSO SVM和PSO SVM方法有更好的识别性能。

Training Method for Support Vector Machine Based on Chaos Particle Swarm Optimization

To make the particles distribute in the problem search space evenly, a novel Chaos Particle Swarm Optimization(CPSO) is proposed based on the analysis of the ergodicity of chaos and inertia weight of the PSO. The improved circle map is introduced, and the new map is introduced into Particle Swarm Optimization(PSO) to avoid PSO from getting into local optimum. A face recognition method using this improved algorithm to train Support Vector Machine(SVM) is presented. Experimental results show that the presented SVM method optimized by CPSO can achieve higher recognition performance.