基于密度估计的社会网络特征簇挖掘方法_韩毅精选.pdf

33 5 Vol.33 No. 5 2012 5 Journal on Communications May 2012 1 1,2 1 1 1 1. 410073 2. 100876 TP311 A 1000-436X(2012)05-0038-11 Mining characteristic clusters: a density estimation approach 1 1,2 1 1 1 HAN Yi , FANG Bin-xing , JIA Yan , ZHOU Bin , HAN Wei-hong (1. School of Computer Science, National University of Defense Technology, Changsha 410073, China 2. Beijing University of Posts and Telecommunications, Beijing 100876, China) Abstract: A hierarchical structure extraction approach based on agglomerative clustering was proposed, and a density esti- mation based on topological structure was designed. By conducting the hierarchical aggregation on layers of hierarchical structure, the characteristic of clusters could be measured. The empirical study conducted on a large real data set indicates that the model and measures are interesting and meaningful, and the algorithms are effective and efficient in practice. Key words: social network; characteristic clusters; data mining 1 [1] [2,3] [5] [4] 2010-12-27 2011-05-25 (6093300560873204) 863 (2010AA012505) Foundation Items: The National Natural Scie