非线性时变PSO优化SVM的入侵检测方法  

Nonlinear Time-varying PSO Optimized SVM Intrusion Detection Method

作　　者：唐风扬 段嘉霖 熊健 覃仁超[1] TANG Fengyang;DUAN Jialin;XIONG Jian;QIN Renchao(School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang 621010)

机构地区：[1]西南科技大学计算机科学与技术学院,绵阳621010

出　　处：《计算机与数字工程》2024年第8期2400-2404,2517,共6页Computer & Digital Engineering

基　　金：四川省科技计划(编号:2022YFG0339)资助。

摘　　要：针对现有使用时变粒子群算法(TVCPSO)优化支持向量机(Support Vector Machines,SVM)对网络流量数据进行入侵检测的方法存在的粒子搜索能力不足等问题,提出了一种非线性时变粒子群算法优化SVM参数的入侵检测方法。该方法首先融合ReliefF算法与信息增益算法对网络流量数据进行特征降维,然后通过非线性学习因子和自适应权重改进时变粒子群算法优化支持SVM,最后通过SVM完成网络流量的检测。NSL-KDD上的结果表明论文方法达到了97.86%的准确率、97.67%的检测率和2%的误报率,验证了方法的有效性。Aiming at the problems of insufficient particle search ability of the existing methods of using time-varying particle swarm optimization(TVCPSO)to optimize support vector machines(SVM)for intrusion detection of network traffic data,an intrusion detection method based on nonlinear time-varying particle swarm optimization(TVCPSO)to optimize SVM parameters is proposed.In this method,firstly,ReliefF algorithm and information gain algorithm are combined to reduce the feature dimension of network traffic data,then the time-varying particle swarm optimization algorithm is improved by nonlinear learning factor and adaptive weight to support SVM,and finally the detection of network traffic is completed by SVM.The results on NSL-KDD show that the proposed method achieves 97.86%accuracy,97.67%detection rate and 2%false positive rate,which verifies the effectiveness of the method.

关 键 词：特征选择 时变粒子群算法 自适应权重 支持向量机 入侵检测 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]