IBeH: Naive Community Detection Methodology for Dark Social Network

In this paper, the authors propose a naive methodology, IBeH, for detecting the community structure of the covert network. Unlike most of the existing methods, IBeH, considers the roles, contribution metric and similarity of the actor profiles. The proposed method computes the reciprocating factor among the actors using the contribution metric rather than the distance metric like density-based. It uses the novel profile diversity analysis to team the actors as a community. This research work also proposes DensCommun, a density based heterogeneous community detection algorithm which is compared with IBeH. The experimental results are derived from the 9/11 dataset. The comparative study significantly presents well-defined communities with minimal overlap using IBeH than DensCommun which inherits the features of the other state-of-art methods like DENGRAPH-IO and evolution in HOCTracker.