Dynamic Behavioral Mixed-Membership Model for Large Evolving Networks
This addresses the challenge of analyzing structural behavior in massive evolving networks like social media or internet traffic, though it appears incremental as an extension of mixed-membership models to dynamic settings.
The authors tackled the problem of modeling node roles in large dynamic networks by proposing a dynamic behavioral mixed-membership model (DBMM) that captures evolving connectivity patterns, with results showing it can be applied to very large networks and uncover interesting dynamic patterns.
The majority of real-world networks are dynamic and extremely large (e.g., Internet Traffic, Twitter, Facebook, ...). To understand the structural behavior of nodes in these large dynamic networks, it may be necessary to model the dynamics of behavioral roles representing the main connectivity patterns over time. In this paper, we propose a dynamic behavioral mixed-membership model (DBMM) that captures the roles of nodes in the graph and how they evolve over time. Unlike other node-centric models, our model is scalable for analyzing large dynamic networks. In addition, DBMM is flexible, parameter-free, has no functional form or parameterization, and is interpretable (identifies explainable patterns). The performance results indicate our approach can be applied to very large networks while the experimental results show that our model uncovers interesting patterns underlying the dynamics of these networks.