A Spectral Analysis Approach for Social Media Community Detection

摘要

Online forums are ideal platforms for worldwide Internet users to share ideas, raise discussions and disseminate information. It is of great interest to gain a better understanding on the dynamic of user interactions and identify user communities in online forums. In this paper, we propose a temporal coherence analysis approach to detect user communities in online forum. Users are represented by vectors of activeness and communities are extracted by a soft community detection algorithm with the support of spectral analysis.