- 期刊
An Improved Particle Swarm Optimization Algorithm Based on Restart Strategy
摘要
In order to solving the problems of premature convergence and local extremum caused by traditional particle swarm optimization algorithm (PSO algorithm), this paper proposes an improved particle swarm optimization algorithm based on restart strategy (BRSPSO algorithm). In the process of optimization, BRSPSO algorithm can update the speed and location of individual particles, increase the influence of excellent particles according to the mutual influence between particles. It introduces a restart strategy, at the same time, to enlarge the search scope of the algorithm and enhance the searching ability for the sake of preventing the search from falling into the local extremum. In this paper, BRSPSO algorithm is used to optimize the parameters of the support vector machine (SVM). Experiments show that compared with SVM optimized by traditional PSO algorithm and SVM without PSO, SVM optimized by BRSPSO algorithm can expand the search space and effectively avoid the local convergence problem of the traditional PSO algorithm.