On Detection of Community Structure in Dynamic Social Networks

Community structure is a very special and interesting property of social networks. Knowledge of network community structure not only provides us key insights into developing more social-aware strategies for social network problems, but also promises a wide range of applications enabled by mobile networking, such as routings in Mobile Ad Hoc Networks (MANETs) and worm containments in cellular networks. Unfortunately, understanding this structure is very challenging, especially in dynamic social networks where social activities and interactions tend to come and go rapidly. Can we quickly and efficiently identify the network community structure? Can we adaptively update this structure based on its history instead of recomputing from scratch?In this chapter, we present two methods for detecting community structures on social networks. First, we introduce

Q

uick

C

ommunity

A

daptation (

QCA

), an adaptive modularity-based method for identifying and tracing the discrete community structure of dynamic social networks. This approach has not only the power of quickly and efficiently updating the network structure by only using the identified structures, but also the ability of tracing the evolution of its communities over time. Next, we present

DOCA

, an quick method for revealing the overlapping network communities that can be implemented in a decentralized manner. To illustrate the effectiveness of our methods, we extensively test

QCA

and

DOCA

on not only synthesized but also on real-world dynamic social networks including ENRON email network, arXiv e-print citation network and Facebook network. Finally, we demonstrate the bright applicability of our methods via two realistic applications on routing strategies in MANETs and worm containment on online social networks.