Network Security Situation Assessment Model Based on GSA-SVM

Abstract

In order to solve the problem of insufficient accuracy caused by the selection of support vector machine (SVM) parameters in situation assessment, a network security situation assessment model (GSA-SVM) based on GSA optimization of SVM parameters is proposed, which combines the characteristics of less parameters needed to be set by gravity search algorithm (GSA) and strong global optimization capability. Firstly, the model receives the situation assessment data evaluated by the expert system, and then searches for the optimal parameters in SVM through GSA, minimize the error between the generated data and the actual network security situation assessment data. The effect of the model is verified by using Eggcrate function and actual situation assessment data. The results show that this method has good learning ability and is better than Particle Swarm Optimization (PSO) in SVM optimization.