Grace Hopper India Annual Conference Schedule

SESSIONS LIST

Social Network Analysis Using Coordination Games

November 16 11:50 am - 12:50 pm
TRACK: Artificial Intelligence, Data Mining & Machine Learning
Jacaranda
Presentation
LEVEL: Intermediate

In social networks like Facebook, Google+, there can be overlap in user’s high-school friends’ circle and his university friends’ circle. It is important to identify the overlapping community structure of a social network, as it helps in understanding the network topology, the spread of information or a rumor in that network. In this paper, we provide a novel, scalable two-phase algorithm NashOverlap to compute the overlapping community structure of a network. To the best of our knowledge, this is the first game theory based scalable overlapping community detection algorithm that can detect an accurate community structure for a given social network. We evaluate our algorithm against the best existing methods for community detection and show that our algorithm improves significantly on benchmark networks with respect to standard normalised mutual information measure.