Assessment Model of Command Information System Security Situation Based on Twin Support Vector Machines

Abstract

In order to solve the difficult issues in Command Information System security situation assessment such as having wide sources of training samples, the time and space complexity is high and the method solving the problems are strong subjective and so on, an approach based on twin support vector machines is proposed. Firstly, the main factors which affect the command information system security are analyzed. Then based on the twin support vector machines, the assessment model of command information system security situation is established. Based on this, the TWSVM is used for sample data learning and parameter estimation, then the global optimum solution is get. Simulation results show that compared with traditional assessment methods, the proposed method can improve the accuracy of the assessment system, reflect the security situation of the current system quickly and efficiently, and has strong robustness.