Research And Implementation Of Network Security Situation Assessment System Based On Machine Learning

Nowadays,with the continuous development of Internet technology,human beings are more and more dependent on the Internet,and the application of Internet technology in various industries has reached a deeper level.However,in addition to the convenience brought by the network to human life,the concern about the security of the network itself has also become the focus of attention of relevant practitioners.Traditional network security protection means have been difficult to cope with today's severe network security situation,in view of this situation network security situation assessment technology application.This technology evaluates the whole network security situation after comprehensively analyzing the information of several main situation elements which are sensitive to the network environment.The application of this technology provides a reference for the next decision of the network administrator.This paper mainly focuses on the establishment of the situation indicator system and the construction of the situation assessment model of network security situation assessment.The specific work of this paper is as follows:(1)By analyzing the selection principles of the network security situation indicator system,a set of hierarchical network security situation indicator system is constructed.The system from the aspects of scientific nature,completeness,independence three considerations,the whole situation index system according to the disaster vulnerability,threat,stability,sex is divided into 4 first-level indicators,indicators under and according to the characteristics of the primary index,divided into 16 secondary indexes,and presents a secondary index in the part of the quantitative formula of nominal data,This situation indicator system can describe the state of the network security environment from whole to part.(2)by using the kernel function in the local kernel function and the characteristics of global kernel function,construct a both local advantages of kernel function and the advantages of global kernel function is mixed kernel function,and introducing the kernel function of hypersphere support vector machine(SVM),to replace the vector inner product operation,make the hypersphere support vector machine(SVM)model has a good global search ability and local search ability,It effectively avoids the shortcoming of traditional support vector machine(SVM)algorithm,which is easy to fall into local optimal solution.Because the traditional support vector machine(SVM)algorithm uses the network search method to optimize the parameters,it has the disadvantage of slow convergence speed.In this paper,an improved cuckoo search algorithm is used to optimize the parameters of the hypersphere support vector machine,so as to construct a network security situation assessment model based on ICS-HSSVM.The experimental results show that this evaluation model has the advantage of higher accuracy(95%accuracy)than other similar models.(3)Design and implementation of network security situation assessment system.Through the data acquisition module of the system,the traffic data in the real network environment can be obtained.By combining the network security situation indicator system constructed in this paper,the relevant situation characteristic values are extracted,and then the network security situation assessment model based on ICS-HSSVM proposed in this paper can be used to obtain the corresponding network security situation assessment values,so as to provide a reference for the next decision of network managers.