Network Situation Awareness Model Prediction Method Based on Genetic Optimization Support Vector Machine

Abstract

Abstract.The support vector machine model is based on the network security situation has strong randomness, is affected by many factors, and the number of types of network security incidents is uncertain, the reference sample is small, the prediction model of need "intelligent", according to SVM forecast algorithm. In order to select the parameters of SVM, genetic algorithm is introduced into the parameter selection in support vector machine, genetic algorithm optimization based on support vector machine structure (GA -SVM) situation awareness prediction model and method to measure data dimensionality reduction using principal component analysis, through simulation experiments, it is proved that this model is the prediction higher precision than neural network, classification and regression tree and cluster analysis prediction model.

Network security situation prediction model based on grey wolf optimization algorithm to optimize support vector machine

“…Wei et al used support vector machine (SVM) model for prediction, which makes up for the deficiencies of the neural network method, but the selection of SVM parameters is blind [1]. Aiming at the selection and optimization of hyperparameters of SVM, Yang et al and Chen et al proposed the use of particle swarm optimization (PSO) and genetic algorithm to optimize [2][3]. Although these methods improve the blindness of SVM parameter selection to a certain extent, they all have different degrees of premature problems, so that the prediction model is not the optimal model.…”