International audience

Detecting communities of interest is a complex problem that has been addressed from different perspectives. In this work, we propose a user-centered approach incorporating social user profiles in community detection for online Social Networks. In our approach, we first compute explicit knowledge acquisition. By exploring the egocentric networks of users, we can infer implicit similarities of interest. The similarity is estimated with reference to homophily and social influence. The latter is leveraged to enhance Sentiment Analysis within communities. Finally, we conduct experiments on datasets extracted from real-world Social Networks.