支持向量机(SVM)的学习性能和泛化能力主要取决于参数选择,然而传统的优化算法难以解决此问题。文中通过支持向量的个数建立优化目标函数,采用微粒群优化(PSO)算法对其优化,寻找最优参数。PSO是一种新兴的基于群体智慧的进化算法。实验表明,微粒群优化算法是支持向量机参数选择的有效方法。
Support vector machine (SVM) has been proved a powerful technique for solving problems in pattern classification and regression, but its learning capacity and generalization capacity mainly depends on the parameters selection of it. Parameters selection for SVM is very complex in nature and quite hard to solve by conventional optimization techniques. So in this paper,a new methodology,based on PSO, for parameters selection of support vector machine is proposed. The simulation (face recognition) result assures the validity of the methodology on time compared with leave - one- out method.