Finding research groups using modularity based community detection algorithm

Co-authorship network is a social network that represents research collaboration between different authors. Each node represents author and an edge can be created between two authors when they have at least one research paper in common. The study of co-authorship networks can reveal interesting patterns of research collaboration such as identifying popular researchers, finding research groups and so on. The focus of this paper is to find research groups in co-authorship networks using modularity based community detection algorithm. The experiments are conducted on two real world co-authorship networks netscience, DBLP. The results are evaluated using two popular metrics in social network analysis namely, modularity and Normalized Mutual Information (NMI). The proposed approach with modularity based community detection algorithm produce good modular research groups in co-authorship networks.