摘要
针对解决网格节点资源聚类问题,提出了基于小生镜遗传算法的模糊聚类分析方法。该算法把小生镜遗传算法搜索的随机性和并行性引入模糊聚类中,对模糊聚类中的聚类中心的个数和聚类中心的选取进行指导,解决了模糊C-均值聚类对初始聚类中心的敏感性问题。实验证明该方法能具有全局收敛性,克服了FCM算法可能陷入局部极小值,并有效地对网格节点资源整合归类,从而改善网格节点资源发现的性能。
This paper presents a method of fuzzy clustering based on niche genetic algorithm in order to solve the clustering problem of grid nodes resource. The method inducts niche genetic algorithm into fuzzy clustering by using its searches randomly and parallelism, which instructs to choose the number of cluster centers and data that are cluster centers. It resolves the problem on sensitiveness of the initial condition of fuzzy C means clustering. Experiment results show that the method has global convergence, avoids local minimum value, sorts effectively grid nodes resource and improve performance of grid resource discovery.
出处
《武汉理工大学学报》
EI
CAS
CSCD
北大核心
2006年第10期101-103,共3页
Journal of Wuhan University of Technology
基金
国家自然科学基金项目资助(70572079)
关键词
网格节点
小生境遗传算法
模糊聚类
模糊C-均值聚类
niche genetic algorithm
fuzzy clustering
grid Node
fuzzy c-means clustering algorithm