Inferring overlapping community structure with degree-corrected block model

Recent research has shown great interest in statistical inference methods for community detection, not only in models and algorithms but also in the detectability. In this paper we propose a fast community detection algorithm based on the degree-corrected block model. By introducing a parameter to select the candidate solutions, our algorithm is able to detect overlapping communities. Experiments on a range of networks have achieved state-of-the-art results. Moreover, we show that the algorithm based on the degree-corrected block model also suffers the detectability limitation, which is in accord with the most recent research on the detectability threshold.