The Network Situation Predicting Model Based On The PSO-SVM

At this stage, the information level in people’s work and life have occupied a high position, at the same time, network security issues are also becoming prominently. In order to grasp of network security situation timely and accurately, the workers and scholars pay their attention to the network situation prediction. In recent years, researchers have been developed a variety of network situation Prediction methods by extensive practice and exploration.This paper proposes network situation predicting model based on the PSO-SVM.The main work as follows:(1) We statistic nearly three years network security incidents, and then detailed analysis the current situation of China’s network security situation. According to the actual network conditions in China, we explore a new data sample, enrich the factors of network situation and then we build our own data sample library.(2) This paper improve predict performance and accuracy by the particle swarm optimization algorithm applied to the parameter selection of support vector machine.(3) According to large number of experimental analysis and comparison of a variety of forecasting models, we verify the accuracy and performance of the prediction model.This paper established a new data sample library; it not only enriches the factor of network situation, but also improves the accuracy and speed of the network situation prediction. The experiment result shows that this model can give a basic objective, accurate and timely predictions on basic Situation of network security.