融合混沌灰狼优化算法的K-均值聚类算法  

K-Means Algorithm Merged with Chaotic Grey Wolf Optimization Algorithm

作　　者：李金 李素[1] 王祖荣 姜缘平 LI Jin;LI Su;WANG Zu-rong;JIANG Yuan-ping(School of Computer Science and Engineering,Beijing Technology and Business University,Beijing 100048,China)

机构地区：[1]北京工商大学计算机学院,北京100048

出　　处：《计算机仿真》2024年第5期343-347,358,共6页Computer Simulation

基　　金：国家自然科学基金青年项目(42101470)。

摘　　要：针对K-均值聚类(K-means)算法对初始聚类中心位置敏感、容易陷入局部最优的缺点,提出一种融合混沌灰狼优化算法的K-means算法。算法利用混沌系统的随机性和遍历性产生分布较均匀的Tent混沌序列来初始化灰狼种群,获得分布较均匀且多样性较高的初始解,提高了算法的全局搜索能力;在搜索全局最优聚类中心的过程中引入基于精英个体的变异操作,维持种群的多样性,提高了算法的的局部搜索能力,避免算法陷入局部最优。实验结果表明,与基本K-means、基于粒子群算法(PSO)改进的K-means、传统灰狼优化算法(GWO)改进的K-means相比,以上算法具有更优的聚类效果,更强的寻优能力。The K-means clustering algorithm is sensitive to the location of the initial clustering center and is easy to trap into the local optimal.To overcome these disadvantages of the K-means algorithm,a K-means algorithm merged with a chaotic grey wolf optimization algorithm is proposed.The features of chaotic randomness and ergodicity are applied to generate a uniformly distributed Tent chaotic sequence for initializing the grey wolf population,achieving uniformly distributed initial solutions and high diversity of population,which can enhance the global search ability.In the process of searching cluster centers of the global optimum,elite-based mutation operator is applied to maintain the diversity of the population,which can enhance the local search ability and avoid trapping into the local optimal.Compared with K-means,PSO,and GWO,the experiment results show that the proposed algorithm has a better cluster effect and stronger optimization ability.

关 键 词：灰狼优化算法 混沌序列 全局最优 局部最优 变异 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]