Incremental Evolutionary Community Discovery Method Based on Neighbor Subgraph

With the popularity of online social networks, dynamic community discovery has received more and more attention. The existing dynamic community discovery algorithm divides the network into multiple time slices and uses the static community detection method to identify the community on each network, but its efficiency is low. How to effectively handle changes in the network structure at different periods, is the key to analyse the dynamic community evolution. Aiming at the form of dynamic network data flow, this paper based on the neighbor subgraph and put forward an incremental community discovery algorithm, which limits the influence of the increment on the network to the neighbor subgraph of the incremental node. The core nodes in the network are found through the local information of the nodes, and the update strategy is designed according to the incremental type and the type of the involved nodes to update the structure of community. Analyse the evolution of the community structure and provide dynamic community structure in real-time as the community structure changes as incremental events occur. Finally, experiments are conducted on static social networks and dynamic social network. Experimental results verify the superiority and efficiency of the algorithm.