Detecting (k, r)-Clique Communities from Social Networks.

Fei Hao,Liang Wang,Yifei Sun,Doo-Soon Park
DOI: https://doi.org/10.1007/978-981-13-1328-8_75
2018-01-01
Abstract:Social communities' detection in large social networks is playing an effective role for analyzing users' behavior and activities over the social media. K-clique community, as one of important topological structure for social networking analysis, provides an efficient coarse knowledge granularity for understanding the structure of the social networks. However, the k-clique community may be lack of cohesiveness because the hops of two nodes in a community may be very large. To this end, a novel structure called (k,r)-clique community is studied. Further, we adopt formal concept analysis approach for exploiting the formation principle of (k,r)-clique community. Consequently, an efficient detection approach for identifying the (k,r)-clique communities is put forwarded.
What problem does this paper attempt to address?