Bat algorithm for overlapping community detection

In social network, a group of elements sharing common interests is called community. To know the structure of these communities, many works have been proposed with different techniques; we can cite node based methods and link based methods. This structure is complex where communities overlap every time. In this paper we use bat algorithm to discover overlapping communities. Bat algorithm is a novel metaheuristic characterized by the echolocation behavior of bats. The algorithm we propose in this paper is based on the links of the network. The objective function evaluates the link density which is convenient for overlapping communities. Experiment on real networks show that the communities discovering with our approach have a higher density.