CMI Silver Jubilee Lecture
Dipak K. Dey, Department of Statistics, University of Connecticut, USA
Clique-based Method for Social Network Clustering
Tuesday, January 6, 2015
Many networks in real life are found to divide naturally into small communities. Examples include Facebook, LinkedIn, computer networks, and metabolic network etc. The problem of detecting clusters or communities is of great importance. An effective and commonly used measurement on the quality of a clustering is called “modularity”, and algorithms that maximize this quantity are among the most popular network clustering approaches nowadays. Unfortunately, modularity method has resolution limit when network is large. Here, we propose a novel network clustering algorithm, which provides user control on this limitation. In addition, we provide detailed results of applying our algorithm in various networks. An analysis about the validity of our algorithm is also included.
(Joint with Guang Ouyang)