关键词: 主题关注网络, 集对, 联系度, 社区发现

Abstract: At present, the interest’s social networks, which share interests or topics, lead to a reform wave of social network. A new social network model, the topic-attention model, fusing the social relations and the topic-attention relation is constructed. Based on this model, firstly, using set pair connection degree to define the similarity between vertices, the measurement can better describe network structure characteristics, overcome the under-estimating for some similarity between vertices based on traditional local structures, and reduce the computational complexity of global similarity indices. Secondly, considering the influence of the topic and the social relationships, a new method of community discovery is proposed based on the set pair connection degree and the clustering algorithm. Finally, in Karate network and douban data set, experiment on the topic community mining, the results show, considering the impact of the topic influence has a better community structure.

Key words: topic-attention network, set pair, connection degree, community discovering