Research on User Behavior Predicting Algorithm Based on Game Theory and Information Fusion Theory

In order to solve the problem of internal network threats, the existing solutions were analyzed based on model, graph theory and access control algorithm. Most of them are depended upon traditional intrusion detection system which are impacted by false positive rate and not suitable for insider problem. In this paper, we proposed our algorithm for insider threat situation awareness, which is based on game theory and information fusion. We use DBN structure and exact inference to acquire and fuse different type of insider information for behavior analysis. As a result of simulation experiment, the algorithm can predict the behavior trend of the internal threat, obtain situation awareness, and solve the internal threat problem. The algorithm has good convergence performance and accuracy.