摘要
针对K均值聚类算法存在的缺点,提出了一种新的聚类算法———基于粒子群的K均值聚类算法,并将此算法与现有的基于遗传算法的K均值聚类算法进行比较.理论分析和数据实验证明,该算法有较好的全局收敛性,不仅能有效地克服传统的K均值算法易陷入局部极小值的缺点,而且全局收敛能力优于基于遗传算法的K均值聚类算法.
After analyzing the disadvantages of the classical K-means clustering algorithm, this paper proposes a novel K-means clustering based on Particle Swarm Optimization algorithm and compares it with Genetic clustering algorithm. The theory analysis and experimental results show that the algorithm not only avoids the local optima, but also has greater searching capability than the existing genetic clustering algorithm.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2005年第6期54-58,共5页
Systems Engineering-Theory & Practice
