Optimal Reducing the Solutions of Support Vector Machines Based on Particle Swam Optimization

Abstract

SVM has been winning increasing interest in areas ranging from its original application in pattern recognition to other applications such as regression analysis due to its noticeable generalization performance. One of the SVM problems is that SVM is considerably slower in test phase caused by large number of the support vectors, which greatly influences it into the practical use. To address this problem, we proposed an adaptive particle swarm optimization (PSO) algorithm which based on a reasonable fitness to optimally reduce the solutions for an SVM by selecting vectors from the trained support vector solutions, such that the selected vectors best approximate the original discriminant function. Experimental results show that adaptive particle swarm optimization algorithm is an optimal algorithm to simplify the solution for support vector machines and can reduce the solutions for an SVM by selecting vectors from the trained support vector solutions.