|
基于SVM的机器学习工业信息IDS攻击性能分析
陈军 2026/5/21 20:31:01
陈军
江苏省高淳中等专业学校（高淳开放大学），江苏南京 211300
【摘要】为了进一步提高机器学习的能力，设计了一种基于支持向量机（support vector machine, SVM）的机器学习网络工业信息入侵检测系统（intrusion detection system, IDS）攻击性能分析方法。工业信息入侵检测系统分成训练和检测流程，在训练中根据已有正常样本和异常样本进行训练，在检测中使用支持向量机分类器来完成最后的识别过程。研究结果表明：通过测试确定最优的参数：罚因子为0.1，搜索步进为0.7，此时系统更易形成平稳收敛效果。各类攻击在SVM学习阶段的准确率出现了较大波动，只出现少量误判的情况，能够获得优秀的攻击检测能力。该研究有助于提高工业信息的网络安全水平，为后续工艺优化奠定一定的理论基础。
【关键词】机器学习；支持向量机；入侵检测；攻击性能
中图分类号：TP309
Analysis of Attack Performance of Machine Learning Industrial Information IDS Based on SVM
Chen Jun
Jiangsu Gaochun Secondary Vocational School (Gaochun Open University), Nanjing, Jiangsu 211300
Abstract: In order to further improve the ability of machine learning, an attack performance analysis method for the industrial information intrusion detection system (IDS) based on the machine learning network of support vector machine (SVM) is designed. The industrial information intrusion detection system is divided into training and detection processes. During training, it is trained based on existing normal and abnormal samples. In detection, a support vector machine classifier is used to complete the final recognition process. The research results show that by determining the optimal parameters through testing: the penalty factor is 0.1 and the search step is 0.7. At this time, the system is more likely to form a smooth convergence effect. The accuracy of various attacks fluctuated significantly during the SVM learning stage, with only a few misjudgments occurring, and excellent attack detection capabilities could be achieved. This research is conducive to enhancing the network security of industrial information and laying a certain theoretical foundation for subsequent process optimization.
Key words: Machine learning; Support Vector Machine; Intrusion detection; Attack performance
0 引言
入侵检测技术（intrusion detection system, IDS）是通过收集网络或系统关键节点，分析是否存在违反安全策略行为或攻击迹象的安全技术。近年来，随着机器学习技术的持续进步（未完，下一页）
附件下载：基于SVM的机器学习工业信息IDS攻击性能分析
|