A Fast Method of Detecting Overlapping Community in Network Based on LFM

Detect overlapping communities efficiently and effectively in various social networks has been more and more important. Aiming at the high complexity of expanding strategy and the defect of generating many homeless nodes in LFM, we propose a quick algorithm based on local optimization of a fitness function(QLFM). The proposed algorithm firstly select a node as seed randomly .With a local fitness function ,the algorithm then will expand from inside to outside of the seed according to the Breadth-First-Search in graph. As different seeds will expand to different communities independently ,and these communities have same nodes ,thus our method can detect overlapping nodes quickly and efficiently. An empirical evaluation of the method using real and synthetic datasets shows that the method give better result not only in time efficiency, but also in quality aspect than other methods at the overlapping community detection.