摘要
传统机器学习算法需要人工构建样本特征,处理海量多源异构网络入侵数据时分类效果较差。针对该问题,结合深度信念网络(DBN)和核极限学习机(KELM),提出一种混合深度学习入侵检测算法DBN-KELM。利用DBN提取高维网络历史数据的抽象特征,获得原始数据的低维表示形式。在此基础上,通过KELM对低维表示的数据做监督学习,达到准确识别网络攻击的目的。在NSL-KDD数据集上进行仿真,实验结果表明,DBN-KELM算法能够提高分类准确率,降低对小样本攻击的误报率,同时缩短分类器的训练时间。
Traditional machine lesming algorithms need to construct sample features manually,which have poor clasification efectwhen dealing with ma sive multi-source intrusion datain heterogeneous network.To solve this problem,a hybrid deep lesrning intrusion detection algorithm is proposed combing Deep BelieC Network(DBN)with Kernel Extreme Learning Machine(KELM),which is named DBN-KELM.It uses DBN to extract the abstract features of high historical data in dimensional network,so as to obtain the low dimensional representlion form of the original data.On this basis,O uses KELM to do supervised learning for low dimensional data to accurately identify the network attack.Simultions9rec9ried outon theNSL-KDD d9tset,9nd theexperimentlresultsshow th9t,DBN-KELM 9lgorithm c9n improve the 9ccur9cy of cl ssific9tion,reducethef9lse9lrm r9te of sm9l s9mple 9tcks 9nd shorten the tr9ining timeof thecl sifier.
作者
汪洋
伍忠东
火忠彩
WANG Yang;WU Zhongdong;HUO Zhongcai(School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2019年第10期171-175,182,共6页
Computer Engineering
基金
甘肃省高等学校创新团队项目(2017C-09)
兰州市科技局科技项目(2018-1-51)
关键词
深度学习
深度信念网络
特征提取
核极限学习机
入侵检测
deep learning
Deep Belief Network(DBN)
feature extraction
Kernel Extreme Learning Machine(KELM)
intrusion detection