Risk Prediction Method of information system based on Bayesian Game

The risk of information system is influenced by both attackers and defenders, so it is reasonable to consider the behaviors of both sides. In this paper a Risk Prediction Model based on Bayesian Game (RPM-BG) is proposed. And an improved revenue calculation method is presented, which analyzes the counterattack and takes the cost parameters and benefits parameters into account at mean time, therefore the revenue could be calculated more accurately. According to the revenue matrices, attacker´s strategy preference can be credibly predicted based on the hybrid strategy of Nash Equilibrium. The results have certain guiding significance for information system security protection. The example analysis proves the effectiveness of the model and the method.