Isabel Kloumann
In this talk we will think about a variety of real world social networks that are comprised of nodes with interesting and diverse properties. We think about groups of nodes that share common features, and reason about group membership identification, diffusion in the network, and overall longevity and dynamics.
In the first part of this talk we have a social network of people and would like to know how to best identify members of an interesting but unlabeled group. Given a few exemplar members of the group how can the network be used to find the hidden remaining ones? This is the seed expansion problem in community detection, and a growing body of work has used seed expansion as a scalable means of detecting overlapping communities. We evaluate several variants of seed expansion and uncover subtle trade-offs between different approaches. We also explore topological properties of communities and seed sets that correlate with algorithm performance, and explain these empirical observations with theoretical ones.
In the second part we will discuss another social network -- Facebook. Facebook Login is a global platform for app usage, and this work studies the lifecycles of a collection of popular Facebook Login apps. App users are groups of nodes that exist within this social ecosystem in a dynamic way. At the temporal level we develop a retention model that represents a user's tendency to continue using an app, and find that a very small parameter set characterizes this dynamic process. At the social level we organize apps along two fundamental axes -- popularity and sociality -- and show how a user's probability of adopting an app depends on properties of both the local network structure and the match between the user's attributes, their friends' attributes, and the dominant attributes within the app user population. We show how our models give rise to compact sets of features with strong performance in predicting the longevity of an app's success.