基于WOA-XGBoost模型的网络入侵检测  

Network Intrusion Detection Based on WOA-XGBoost Model

作　　者：闫海涛 张之义[1] 朱晓明 王鹏[1] YAN Haitao;ZHANG Zhiyi;ZHU Xiaoming;WANG Peng(China Electronics Technology Group Corporation No.54 Research Institute,Shijiazhuang 050081,China)

机构地区：[1]中国电子科技集团公司第54研究所,石家庄050081

出　　处：《计算机测量与控制》2023年第3期127-133,共7页Computer Measurement &Control

摘　　要：网络入侵检测系统(NIDS)是检测网络攻击和维护网络安全的关键技术之一,是网络安全领域中的重要研究方向;近年来,研究者利用机器学习算法来完成入侵检测任务并取得了很好的成果,但检测效率和精确率有待进一步提升;在对鲸鱼优化算法(WOA)和极限梯度提升算法(XGBoost)的特点进行实验和对比分析的基础上,提出了WOA-XGBoost模型,首先构建基于XGBoost的分类模型,然后利用WOA算法自适应搜索XGBoost的最优参数,最后基于NSL-KDD数据集评估所提出WOA-XGBoost模型的性能;实验结果表明,该模型在分类精确率、准确率、召回率和AP指标方面均优于其他模型如XGBoost、随机森林、Adaboost和LightGBM;该工作也为群体智能优化算法在网络入侵检测中的应用提供了依据。Network intrusion detection system(NIDS) is one of the key technologies to detect network attacks and protect network security, it is an important research direction in the field of network security. In recent years, machine learning algorithms are used to complete intrusion detection tasks and achieve good results, but the detection efficiency and accuracy need to be further improved. Based on the experiments and comparative analysis of the whale optimization algorithm(WOA) and extreme gradient boosting algorithm(XGBoost), a WOA-XGBoost model is proposed. Firstly, a classification model based on the XGBoost is constructed, then the WOA algorithm is used to search the optimal parameters of the XGBoost adaptively. Finally, the performance of the proposed WOA-XGBoost model based on the NSL-KDD dataset is evaluated. Experimental results show that the classification precision, accuracy, recall and AP indicators of the model are better than that of other models such as XGBoost, Random Forest, Adaboost and LightGBM, it provides a basis on the application of swarm intelligence optimization algorithm in network intrusion detection.

关 键 词：网络安全 入侵检测 异常行为检测 WOA-XGBoost 集成学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]