A Novel Virus Detection and Active Defense Algorithm Based on SVM Optimized by Differential Evolution Algorithm

Abstract

This paper proposes a novel active defense strategy focuses on users' behavior patterns which to classify the behaviors accurately by SVM for virus detecting. Differential evolution was introduced to improve the precision of SVM and turns it into an optimization problem which object is the classification precision. And the parameters are regarded as the variables to be optimized. The experimental results show that the proposed model has a higher precision than the compared methods, such as BPNN, SVM, GA-SVM, etc. In addition, the method is more efficient so that, it can be quickly updated and applied.

Research on the game of network security attack‐defense confrontation through the optimal defense strategy

“…Wang et al 7 established a static Bayesian game model, designed a method to choose the active defense strategy based on the types of attack and defense, calculated and analyzed the method, and verified the methods' performance by an example analysis. Ai et al 8 designed an active defense method for user behavior mode, achieved the accurate classification of virus behavior based on support vector machine and combining with differential evolution, and found that the method had high accuracy and could be updated and applied quickly. Zhang et al 9 established a cloud‐based active defense system for malicious code in the network, collected threat data through the honeypot system, and conducted the intelligent analysis of program behavior through multiple servers, forming a comprehensive and intelligent active defense system.…”