Analyzing the Co-evolution of Network Structure and Content Generation in Online Social Networks

Abstract

With the rapid growth of online social network sites (SNS), it has become imperative to investigate what drives content production on these platforms. We posit that the content producing behavior of users is influenced not just by their personal attributes like age and gender, but also by their social network structure. However, it is empirically challenging to estimate network structure and behavior through traditional approaches as the social network structure and the content production behavior influence the evolution of each other. In the current study, we adapt an actor-based continuous-time model to jointly estimate the co-evolution of the users' social network structure and their content production behavior using a Markov Chain Monte Carlo (MCMC) based simulation approach. We apply our model to an online social network of university students and uncover strong evidence for both social influence and homophilous friend selection. Interestingly, we find that individuals befriend others who are similar in content production during the friendship formation stage, but gradually diverge in their content production behavior from these similar others over time. We offer potential explanations for this phenomenon and emphasize the importance of these findings for platform owners and product marketers.