Query-Oriented Active Community Search

Online social network (OSN) is now one of the emergent platforms to make social connections and share individual’s ideas, feelings, opinions, etc. Community detection in OSN is a hot research topic. Most of the existing research works on finding local community mainly concentrate on properties or attributes of the social network users or rely on the network topology. However, not enough attention is paid to the users’ interests or activeness on different attributes, which play a key role for forming an online community. As a result, the resulting communities may have users with diverse interests, as well as different degrees of activeness on different attributes. In this work, our aim is to detect attribute-oriented active community search in OSN, that is, for a given input query consisting a query node (user) and a set of attributes, we want to find densely connected community in which community members are actively participating with respect to the given query attributes. We propose a novel attribute relevance activeness score function for the candidate community members to search the desired active community. We conduct extensive experiments on two real data sets to demonstrate the effectiveness of our proposed method and also report some interesting observations.