Meetup Group Life Cycle Study

Meetup is online social networking portal that facilitates offline group meetings. Since 2001, there have been over 500,000 groups created on meetup, with totally over 9.3 million users registered, over 15.2 million events organized. It has witnessed a booming growth during recent years. However, among all the groups created, only part of them can stay active and vibrant for a long time while others became stagnant or dead after a while. An astonishing fact is that actually 80% of the 500,000 groups are already inactive. This pose questions to both meetup and those group owners what cause group to die? Can we predict life and death in advance thus take early actions? Or are there any patterns of the group life cycle? This paper is aimed to answer these questions. In this paper, we first studied about the key factors that influences the group longevity. Then we used different machining learning algorithm to predict the group life and death. Then, we analyzed the group life cycle based the group growth curve and proposed explanations to some key questions. The rest of the paper is organized as follows: In Part II, our work on group life and death prediction is presented in details. In Part III, we talk about life cycle analysis and our hypothesis. Conclusions are summarized in Part IV.