Prediction model of network security situation based on genetic algorithm and support vector machine

Abstract

In order to predict the development trend of network security situation more accurately, this paper proposes an improved vector machine model by simulated annealing optimization to improve network security situation prediction. In the process of prediction, the sample data of phase space reconstruction network security status is first formed to form training sample set, and then the simulated annealing method is improved. The correlation vector machine is the optimization of correlation vector machine with simulated degradation algorithm embedded in the calculation process of objective function. The network security situation prediction model is obtained through super parameters to improve the learning ability and prediction accuracy. The simulation results show that this method has higher prediction accuracy better than the correlation vector machine model optimized by Elman and simulated annealing. This method can describe the change of network security well.