Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (33): 132-134.DOI: 10.3778/j.issn.1002-8331.2009.33.043

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Clustering method based on improved particle swarm optimization

SUN Yang,LUO Ke   

  1. Institute of Computer and Communication Engineering,Changsha University of Sciences and Technology,Changsha 410076,China
  • Received:2008-07-02 Revised:2008-09-25 Online:2009-11-21 Published:2009-11-21
  • Contact: SUN Yang

基于改进的粒子群算法的聚类算法

孙 洋,罗 可   

  1. 长沙理工大学 计算机与通信工程学院,长沙 410076
  • 通讯作者: 孙 洋

Abstract: This paper proposes a clustering algorithm which is based on improved Particle Swarm Optimization(PSO).Both the K-means,which has strong capacity of local searching,and the cross,mutation operation,which are based on the genetic algorithm,are combined in the PSO algorithm.It not only improves the PSO’s local searching capacity,accelerates the convergence rate,and effectively prevents the premature convergence,for it adds cross and mutation operations.Experiments show that this clustering algorithm has a better convergence.

Key words: cluster analysis, particle swarm optimization algorithm, K-means, genetic algorithm

摘要: 提出了一种基于改进的粒子群算法的聚类方法。该算法是将局部搜索能力强的K-均值算法和基于遗传算法的交叉、变异操作同时结合到粒子群算法中。既提高了粒子群算法的局部搜索能力、加快了收敛速度,同时因为加入了交叉、变异操作,有效地防治了早熟收敛现象的发生。实验表明该聚类算法有更好的收敛效果。

关键词: 聚类分析, 粒子群算法, K-均值算法, 遗传算法

CLC Number: