基于深度学习的计算机网络入侵行为检测  

Computer Network Intrusion Behavior Detection Based on Deep Learning

作　　者：袁玮 李勇 YUAN Wei;LI Yong(Hengshui Human Resources and Social Security Bureau,Hengshui,Hebei 053299,China;Hengshui Jizhou District Integrated Media Center,Hengshui,Hebei 053299,China)

机构地区：[1]衡水市人力资源和社会保障局,河北衡水053299 [2]衡水市冀州区融媒体中心,河北衡水053299

出　　处：《移动信息》2025年第1期136-138,147,共4页Mobile Information

摘　　要：为提高计算机网络的安全性,实现对网络异常行为的精准检测,文中结合深度学习技术,对计算网络入侵行为检测方法展开了全面的设计与研究。计算机网络用户行为数据可分为定量与定性两类,通过归一化处理、平衡处理、降维处理,可完成数据预处理;利用深度学习中的CNN模型进行数据训练,可实现用户异常行为特征提取;应用支持向量机,利用最优超平面,分隔不同类别的数据,进行网络入侵行为的精准辨识与检测。实验结果表明,文中设计的方法不仅可以实现对网络异常行为类别的准确辨识,还可以实现对异常行为数据的精准检测。In order to improve the security of computer networks and achieve accurate detection of abnormal network behavior,this paper combines the application of deep learning technology to comprehensively design and research intrusion detection methods for computing networks.Computer network user behavior data is divided into quantitative and qualitative categories,and data preprocessing is completed through normalization,balancing,and dimensionality reduction;Using CNN models in deep learning to train data and extract abnormal user behavior features;Applying support vector machine,utilizing the optimal hyperplane to separate data of different categories,for precise identification and detection of network intrusion behavior.The comparative experimental results show that the method designed in this paper can not only accurately identify the categories of network abnormal behavior,but also achieve precise detection of abnormal behavior data.

关 键 词：深度学习 数据预处理 特征提取 检测方法 入侵行为 计算机网络 

分 类 号：TN271[电子电信—物理电子学]