Users Recommendation Based On Formal Concept Analysis In Social Network

With the emergence of Web 2.0, micro-blog systems, such as Twitter and Tencent Weibo, have attracted more and more users’ attention. Different from traditional online social network, in the community of micro-blog system, users are following or followed by theirs followees or followers, which forms a social relationship. As a platform for information sharing and propagation, the most fundamental and unique service provided by micro-blog system is that it can be used as a tool for users to post short messages(called as micro-blog) whatever they want. In every user’s homepage, the recent micro-bolg from all his/her followees are listed chronologically. As to followee recommendation in micro-blog systems, it is thus foremost to recommend most relevant followees to users so that they can benefit most from the micro-blogs information gathered from their followees.We show how Formal Concept Analysis(FCA) can be applied to Friends Recommender in social network. FCA is a mathematical method for analyzing binary relations. Here we apply it to the relation between users and terms of micro-blogs text in a friends recommender system. This paper carries through a useful explore and research, the two main points of the research as follows:(1) The concept lattice was utilized to store the knowledge context based on the relationship among users and terms in order to guide a recommender for the recommending. By computing the concept similarity and matching visited candidate users with the concept lattice, we obtained the rank values of the candidate users to expand the target users’ friend lists.(2) We propose the UCG(user context graph) to represent the knowledge context based on the user’s twitter and social interactions to recommend friends. By computing the concept similarity among the concepts in the concept context graph and matching users with the concept context graph, we compute the rank values of the target users for recommendation. Experimental evaluation was conducted in order to determine the impact of concept clustering means based on UCG. Experimental results show that the proposed methods is satisfactory for effective and efficient friends recommendation.