An Approach for Opinion-Demographic-Topology Based Microblog Friend Recommendation

摘要: Through the tremendous increase of users on microblogging social networks with their associated streams content, scarcity one user’s attention arises. The process filtering such massive content and discovering who other could be aligned his own interests would consume much time. Thus, various mechanisms have been investigated to recommend friends by analyzing posted graph, or user profiles. In this paper, we propose a new approach for microblog friend recommendation based opinion, sentiment, towards topics in microblogs combined addition demographic data available profiles, including age, gender, location. We deployed cloud-based recommender service using R language big analytics, which applies our proposed gather feedback from real Twitter users. Results show 0.77 average precision value, 21 % rate considering opinion mining.