基于局部聚合的复杂网络自动聚簇算法  

An Automatic Clustering Algorithm for Complex Networks based on Local Agglomeration

作　　者：汤蓉[1] 唐常杰[2] 徐开阔[1] 杨宁[2] 

机构地区：[1]成都信息工程学院计算机学院,成都610225 [2]四川大学计算机学院,成都610065

出　　处：《电子科技大学学报》2014年第3期329-335,共7页Journal of University of Electronic Science and Technology of China

基　　金：国家自然科学基金(60773169);国家"十一五"科技支撑计划(2006BAI05A01)

摘　　要：针对全局聚簇计算消耗偏高的缺陷,提出了基于局部聚合的自动迭代聚簇算法(LAICA)。LAICA首先通过局部聚簇发现网络中连接紧密的节点集,即:局部簇,然后迭代合并局部簇从而实现对网络的全局聚簇。实验结果表明,LAICA算法能自动决定簇数并精确分配节点至其所属簇,在真实网络数据集上的聚簇精确度最高达到99.72%。By adopting local clustering, an automatic iterative clustering algorithm（LAICA） is proposed for complex networks. LAICA first agglomerates groups of nodes tightly connected as local clusters, and then iteratively merge local clusters based on some global modularity criterion. The experiment results demonstrate that LAICA can not only resolve the number of communities automatically, but also assign nodes into their communities accurately. The highest fraction of vertices classified correctly reaches 99.72% on real benchmark networks.

关 键 词：复杂网络 全局聚簇 局部聚簇 局部模块性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]