Community structure discovery and community topic analysis in microblog

With the rise of media like microblog, discovering community and analysing the characteristics of network in the microblog network have gradually became a hotspot of research in the field of social network analysis in recent years. In this paper, based on the property contents of microblog users namely users' interest, and considering the structural similarity and attribute similarity of cliques got from Clique Percolation Method (CPM algorithm), we improved the CMP algorithm from the perspective of its output by mergering the cliques. This improvement resolves the problems that the definition of clique by CPM algorithm is too strict and does not meet the reality. Then, our research analysed the topic of the communities found by the improved CPM algorithm using Fuzzy Comprehensive Evaluation Method. And we obtained communities with a higher application value finally. In the end, we verified our research using the real data crawling from Sina Weibo, a microblog site which is the most popular in China.