基于变异精密搜索的蜂群聚类算法

被引:17
作者
罗可
李莲
周博翔
机构
[1] 长沙理工大学计算机与通信工程学院
关键词
聚类; 粗糙集; 人工蜂群; ?-means; 变异算子;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
140502 [人工智能];
摘要
针对K-means聚类算法过度依赖初始聚类中心、局部收敛、稳定性差等问题,提出一种基于变异精密搜索的蜂群聚类算法.该算法利用密度和距离初始化蜂群,并根据引领蜂的适应度和密度求解跟随蜂的选择概率P;然后通过变异精密搜索法产生的新解来更新侦查蜂,以避免陷入局部最优;最后结合蜂群与粗糙集来优化K-means.实验结果表明,该算法不仅能有效抑制局部收敛、减少对初始聚类中心的依赖,而且准确率和稳定性均有较大的提高.
引用
收藏
页码:838 / 842
页数:5
相关论文
共 15 条
[1]
函数优化的蜂群算法 [J].
刘勇 ;
马良 .
控制与决策 , 2012, (06) :886-890
[2]
The best-so-far ABC with multiple patrilines for clustering problems[J] Anan Banharnsakun;Booncharoen Sirinaovakul;Tiranee Achalakul Neurocomputing 2013,
[3]
A combined approach for clustering based on K -means and gravitational search algorithms[J] Abdolreza Hatamlou;Salwani Abdullah;Hossein Nezamabadi-pour Swarm and Evolutionary Computation 2012,
[4]
A global best artificial bee colony algorithm for global optimization[J] Weifeng Gao;Sanyang Liu;Lingling Huang Journal of Computational and Applied Mathematics 2012,
[5]
Efficient Clustering Method Based on Rough Set and Genetic Algorithm[J] Jianyong Chen;Changsheng Zhang Procedia Engineering 2011,
[6]
A modified artificial bee colony algorithm[J] Wei-feng Gao;San-yang Liu Computers and Operations Research 2011,
[7]
A clustering method combining differential evolution with the K -means algorithm[J] Wojciech Kwedlo Pattern Recognition Letters 2011,
[8]
An efficient hybrid approach based on PSO; ACO and k -means for cluster analysis[J] Taher Niknam;Babak Amiri Applied Soft Computing Journal 2009,
[9]
An artificial bee colony approach for clustering[J] Changsheng Zhang;Dantong Ouyang;Jiaxu Ning Expert Systems With Applications 2009,
[10]
A quantum-inspired genetic algorithm for k -means clustering[J] Jing Xiao;YuPing Yan;Jun Zhang;Yong Tang Expert Systems With Applications 2009,