Social Status and Communication Behavior in an Evolving Social Network

The degree to which individuals can exert influence on propagation of information and opinion dynamics in online communities is highly dependent on their social status. Therefore, there is a high demand for identifying influential users in a community by predicting their social position in that community. Moreover, understanding how people with various social status behave, can shed light on the dynamics of interaction in social networks. In this paper, I study an evolving online social network originated from an online community for university students and I tackle the problem of forecasting users' social status, represented as their PageRank, based on frequency of recurring temporal sequences of observed behavior, i.e. behavioral motifs. I show that individuals with different values of PageRank exhibit different behavior even in early weeks since the online community's inception and it is possible to forecast future PageRank values given frequency of behavioral motifs with high accuracy.