An improved label propagation algorithm using average node energy in complex networks

•We propose an improved label propagation algorithm (LPA) to uncover overlapping community structure.•The improved LPA can identify the bridge nodes in each iteration and then we can uncover overlapping communities when the iteration terminates.•The introduced algorithm can effectively uncover reasonable overlapping community structures in the real-world and social networks.

Detecting overlapping community structure can give a significant insight into structural and functional properties in complex networks. In this Letter, we propose an improved label propagation algorithm (LPA) to uncover overlapping community structure. After mapping nodes into random variables, the algorithm calculates variance of each node and the proposed average node energy. The nodes whose variances are less than a tunable threshold are regarded as bridge nodes and meanwhile changing the given threshold can uncover some latent bridge node. Simulation results in real-world and artificial networks show that the improved algorithm is efficient in revealing overlapping community structures.