Research On Classification Of Intrusion Detection Based On Support Vector Machine Algorithm

With the rapid development of network technology,the Internet has greatly promoted social development and its application has become more and more widespread.However,the issue of network security has become increasingly prominent.There have been endless intrusions of website malicious code,hacking,and so on.It even poses a huge threat to the security of our country at the national level,so it is very necessary to conduct research on network security technologies.Intrusion detection technology can effectively discover the intrusion behavior by analyzing the key information on the computer system and the network,thus intercepting and responding before the attacker damages the network system.With the increasing scale of networks,the highdimensional features of network data are becoming more and more obvious.Such data can easily lead to a decrease in the learning performance of the classifier.Among various methods of intrusion detection,support vector machine has high detection speed,high detection accuracy and good generalization ability.Therefore,this paper uses support vector machine to establish intrusion detection model.Multi-Verse Optimizer algorithm,a group intelligent algorithm,has the characteristics that it requires few parameters and finds an approximate optimal feasible solution in a relatively short period of time,but it is prone to fall into a local optimum.In response to this shortcoming,this paper proposes Chaotic Multi-Verse Optimizer(CMVO)algorithm by combining it with Chaos theory.It's less likely to fall into local optimum by designing a better initial universe which improves the quality of the universe.When it is trapped in a local optimum,the algorithm is made to jump out of local optimum by using chaotic disturbances.After that,it is applied to the intrusion detection model of support vector machine.For the high-dimensional features of network data,this algorithm is used to select features of the data and remove redundant features.In order to select the parameters of the support vector machine while selecting features,the coding format and fitness function are designed,which ensures the classification accuracy and accelerates the detection efficiency.In this paper,the proposed CMVO algorithm is verified by the benchmark function.The experimental results show that it has good exploitation and exploration capabilities.By using the UNSW-NB15 data set,the performance of the intrusion detection model established by the CMVO is tested.The results show that it can effectively improve the detection accuracy and reduce the false positive rate and false acceptance rate.