The Prediction Model of Online Social Networks' Evolution Based on the Similarity of Community

It is one of the key problems in network research to discover the evolution of community in social network, study the evolution of community and forecast the development trend of community. How to more faithfully reveal the evolution of the community in complex networks has always been the goal pursued by all researchers. Based on the evolutionary prediction model of event-based community, this paper constructs four kinds of event framework models by using community similarity algorithm and edge similarity algorithm, and through the WS small world network simulation data set test. In the premise of improving the accuracy of prediction, it does not significantly reduce the number of community events found, with good predictive performance.