用于网络安全态势预测的粒子群与支持向量机算法研究  

PSO AND SVM FOR NETWORK SECURITY SITUATION PREDICTION

作　　者：孙卫喜[1] Sun Weixi(College of Network Security and Information Technology,Weinan Normal University, Weinan 714099, Shaanxi, China)

机构地区：[1]渭南师范学院网络安全与信息化学院

出　　处：《计算机应用与软件》2019年第6期308-316,共9页Computer Applications and Software

基　　金：陕西省自然科学基础研究计划资助项目(2017JM6110);渭南师范学院自然科学类研究项目(18YKS13)

摘　　要：网络安全已经越来越受到人们的重视,网络安全态势预测作为一种阻隔网络安全威胁的新兴手段受到了学者的广泛关注。针对威胁网络安全的特异性因素,提出一种改进的网络安全态势预测技术。介绍网络安全态势感知预测相关的背景;针对网络安全态势预测过程中存在的时变性与非线性等特征,在分析了支持向量机与改进粒子群算法的基础上,给出一种改进的PSO SVM算法。通过相关仿真实验说明该方法的可行性与实用性。实验表明,使用该预测方法处理先前收集到的网络安全数据,明显提高了网络态势的预测精度,实现了对网络安全威胁的有效防御。Network security has been paid more and more attention by people.The network security situation prediction has attracted extensive attention from scholars as an emerging means to block network security threats.This paper proposed an improved network security situation prediction technology for the specific factors that threatened network security.I introduced the background related to network security situational awareness prediction.Then,based on the characteristics of time-varying and nonlinearity in the network security situation prediction process,an improved PSO-SVM was presented based on the analysis of SVM and PSO.Furthermore,the relevant simulation experiments were given to illustrate the feasibility and practicability of the proposed method.Experiments show that using this prediction method to process the previously collected network security data significantly improves the prediction accuracy of the network situation and achieves effective defense against network security threats.

关 键 词：安全态势 支持向量机 粒子群算法 态势预测 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]