A Framework for Multi-Faceted Analytics of User Behaviors in Social Networks

Abstract

In this paper, we propose a framework for conveniently conducting multi-faceted behavior based analytics in social networks. The breadth and depth of social network behavior analytics could thus be dramatically improved. The ＂breadth＂ refers to the coverage of analytic perspectives catering to the various analytic needs of end users, and the ＂depth＂ refers to the multi-dimensional dynamics of social network evolution. A detailed analysis of Twitter was then presented illustrating (i) the multi-faceted relationship between users based on their co-joining categories and co-spreading tweets with three orthogonal dimensions of affect analyzed simultaneously, i.e., valence, activation, and intention; (ii) the semantic network of Twitter topics based on the relations of the users and the tweets marked; (iii) multi-faceted social dynamics, i.e., the rise and fall of the multiple dimensions over time.