Communities Evolution Analysis Based on Events in Dynamic Complex Network

In the dynamic complex networks, the community structure is constantly changing, such as generation, maintaining, merger, die, fusion, etc. These changes will lead to the dynamic evolution of the structure and morphology in an entire community. Current static community division methods can not be well used to analyze the dynamic network evolution. In this paper, based on events in the community evolution, firstly three evolutionary relationships are defined, as well as the definition of community aggregation and community divergence. Then based on the idea of time sheet and the overlapping community division algorithm, dynamic complex networks are divided. The whole evolution of the dynamic community is portrayed, through the comprehensive analysis of the evolution relationships of the community on the adjacent time sheet. And according to values of community aggregation and community divergence, a great polymerization or large dispersion will be found in the community evolution process, so major events hidden in the data can be revealed.