Semantic Clustering-Based Community Detection in an Evolving Social Network

Classic community detection methods in social networks are usually based on graph clustering algorithms which employ the structural information for group identification. They cluster nodes into groups topologically. These methods count purely on the linkage structure of the underlying social media. However, in many applications, it is possible to take into account content issued by users of social media to guide the clustering process. Messages issued by users may express relations between users/entities, which can be utilized for community detection. In this paper, we propose to detect hidden structures in a social network, uses the semantic information extracted from posts in the social media.