Research On Network Security Situation Intelligence Assessment Method Based On Association Rule Analysis

With the vigorous development of economy and science and technology,computer has long been closely related to everyone’s work,study and life,and the network has also been a blowout development,but at the same time,computer Cyber security(CS)problems and related risks are also gradually emerging,CS is undergoing a severe test at the current stage.Nowadays,the security threats and attack means in the network space are developing towards the direction of diversification,standardization and intelligence,and the network security is facing huge challenges.How to use effective protection means to prevent the network from being attacked has become one of the hot research directions in the field of network security.Therefore,the network security situational awareness technology should be born in this context,and the exploration of this technology is of great significance for the effective defense and protection of information security.Network security situation awareness is to capture and collect network security elements with great influence,analyze the data through data fusion and mining,and play a role in predicting the future development trend of network security situation.This thesis mainly evaluates network security situation through Association rule analysis and Bayesian network,and proposes a frequent pattern tree network security evaluation model based on Association rule analysis(ATA).Conduct an in-depth study of network security situation assessment method based on KDD-99 data set.By using FP-tree algorithm,through the preset support degree,the network traffic characteristics are screened,the associated factors affecting network security in the traffic are mining,and then the current network security situation characteristics are extracted,and then these factors affecting network security are used to learn and train the Bayesian network.Through the trained Bayesian network,the current network security situation is perceived.Finally,the network security situation awareness evaluation table is used to analyze and evaluate the current situation.On this basis,a network security evaluation and optimization method based on particle swarm optimization and Bayes is proposed.By adding particle swarm optimization algorithm,the network traffic characteristics based on FP-tree frequent pattern tree are further optimized to improve its feature selection ability,and then the performance and accuracy of network security situation assessment are improved by optimizing Bayesian networks based on constraint and fractional structure learning.Experimental results show that the average classification accuracy of the proposed model is 97.14%,which is better than the traditional network security situation model.