Title:Understanding the growth and death of online social networks: evidence and model

Abstract:Online Social Networks (OSNs) are multi-billion dollar enterprises. Surprisingly, little is known about the mechanisms that drive them to growth, stability, or death. This study sheds light on these mechanisms. We are particularly interested in OSNs where current subscribers can invite new users to join the network. Measuring the relationship between subscriber activity and network growth of a large OSN over five years, we formulate four hypotheses that together describe the observed OSN subscriber behavior. We then provide a model (and extensions) that simultaneously satisfies all four hypotheses. Our model provides deep insights into the dynamics of subscriber activity, inactivity, and network growth rates, even predicting the existence of five classes of OSNs with respect to their subscriber activity time series. We present subscriber activity data of nearly thirty OSN websites, measured over five years, and show that these observed activities are well described by four of the activity time series predicted by our model. We also show evidence that the fifth time series is rare but also exists in the wild. Finally, we discuss how our model may help describe the growth of social movements such as grass-root organizations.