基于ELM-KNN算法的网络入侵检测模型  

Network intrusion detection based on ELM-KNN algorithm

作　　者：顾兆军[1,2] 李冰[1,2] 刘涛 GU Zhao-jun1,2 , LI Bing1,2 , LIU Tao2(1. information Security Evaluation Center, Civil Aviation University of China, Tianjin 300300, China; 2. College of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, Chin)

机构地区：[1]中国民航大学信息安全测评中心,天津300300 [2]中国民航大学计算机科学与技术学院,天津300300

出　　处：《计算机工程与设计》2018年第8期2412-2416,2421,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(61601467;U1533104);民航科技基金项目(MHRD20140205;MHRD20150233);民航安全能力建设基金项目(PDSA008);中央高校基本科研业务费中国民航大学专项基金项目(3122013Z008;3122013C004;3122015D025);中国民航大学科研启动基金项目(2013QD24X)

摘　　要：针对传统神经网络算法在处理入侵检测问题时易陷入局部极小导致分类正确率不高的问题,提出基于极限学习机(ELM)特征映射的K最近邻(KNN)算法的网络入侵检测模型。利用ELM算法将低维输入空间中复杂线性不可分的样本投影到高维特征空间,使其线性可分,用KNN算法对投影到高维特征空间的样本进行分类,建立入侵检测分类器。采用KDD Cup99数据集的仿真结果表明,相比其它入侵检测方法,基于ELM-KNN算法的入侵检测模型提高了入侵检测正确率。Aiming at the problem that the traditional neural network algorithm is easy to fall into the local minimization when dealing with the intrusion detection problem,a network intrusion detection model of K nearest neighbor（KNN）algorithm based on ELM feature map was proposed.The ELM feature mapping algorithm was used to project the complex linear inseparable samples in the low-dimensional input space into the high-dimensional feature space to make the samples linearly separable,and the KNN algorithm was used to classify the samples which were projected into the high-dimensional feature space to establish a network intrusion detection classifier.The simulation experiment was carried out using KDD Cup99 data set.The results show that the intrusion detection model based on ELM-KNN algorithm improves the accuracy of intrusion detection compared with that of other intrusion detection methods.

关 键 词：入侵检测 极限学习机 K最近邻算法 特征空间 分类问题 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]