Earthquake prediction based on community division

Abstract

In time-space influence domain, two directed weighted earthquake network are structured based on the earthquake number and the maximum magnitude of Southern California. The earthquake prediction method is proposed based on the minimum edge weight. By CNM (community detection method) community division algorithm, the network is divided into several communities and the top 10 communities can be selected according to the number of nodes. Finally, we compare the accuracy of the divided network with the network without community division. The simulation results show that the community division can improve the accuracy of the earthquake prediction.