期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：48
期号：2
页码：762-767
出版社：Journal of Theoretical and Applied
摘要：Proposed the Algorithm of K-means (CPSOKM) based on Chaos Particle Swarm in order to solve the problem that K-means algorithm sensitive to initial conditions and is easy to influence the clustering effect. On the selection of the initial value problem, algorithm using particle swarm algorithm to balance the random value uncertainty, and then by introducing a chaotic sequence, the particles move speed and position in a redefined, thus solving the initial value sensitivity, while the algorithm with overall search capability, but also to avoid the local optimum. The algorithm add acceleration factor and escape factor in order to improve the time efficiency. Experiment result proved that the CPSOKM algorithm has a fast convergence speed, high stability, and good clustering effect.
