摘要
针对模糊C均值聚类算法(FCM)存在对初始聚类中心敏感,易陷入局部最优解的不足,将改进的粒子群聚类算法与FCM算法相结合,提出了一种基于粒子群优化的模糊C均值聚类算法。该算法对粒子群初始化空间及粒子移动最大速度进行优化,同时引入环形拓扑结构邻域,提高粒子群聚类算法的全局搜索能力。对UCI中3个数据集进行仿真实验,结果表明提出的基于粒子群优化的模糊C均值聚类算法相比FCM算法和基本粒子群聚类算法具有更好的聚类效率和准确性。
FCM algorithm is sensitive to initial clustering center and liable to be trapped in a local optimum solution. Combining with the improved PSO algorithm,a fuzzy C-means clustering algorithm based on particle swarm optimization was proposed. The algorithm optimizes the particle swarm initialization space and the maximum velocity of particle,and adopts the ring topology neighborhood. The method improved the global search capability of particle swarm clustering algorithm. The experiment results of UCI data set demonstrate that the proposed algorithm has better clustering validity and accuracy than FCM and particle swarm clustering algorithm.
作者
王宇钢
Wang Yugang(School of Mechanical Engineering and Automation, Liaoning University of Technology, Jinzhou 121000, Chin)
出处
《信息技术与网络安全》
2018年第8期36-39,44,共5页
Information Technology and Network Security
基金
辽宁省自然科学基金资助项目(20170540445)
关键词
聚类
粒子群优化
模糊C均值聚类算法
粒子群聚类算法
clustering
particle swarm optimization
fuzzy C-means clustering algorithm
particle swarm clustering algorithm
