摘要
为了保障网络环境的安全性,提出了基于集成式图卷积神经网络算法的网络入侵检测技术。研究方法采用随机梯度下降算法和均方根传播(Root Mean Square Propagation,RMSProp)优化器提升了检测模型的训练效率,强化了检测模型的分类效果。研究结果显示,研究模型的入侵检测准确率为96.41%~97.18%。可见经过研究模型优化后,入侵检测技术在模型训练效率和模型训练精度上都有明显提升。研究模型可以根据访问来源进行数据分类,提升了入侵检测模型对访问行为的分类效果。同时,分类效果的提升优化了计算机对攻击行为的识别效率,使计算机的防御效果增强,有效保障了用户的网络安全环境。因此,研究为网络入侵行为的检测提供了一个识别效果较好的技术方法。
In order to ensure the security of the network environment,a network intrusion detection technology based on integrat-ed graph convolutional neural network algorithm is proposed.The research method uses stochastic gradient descent algorithm and Root Mean Square Propagation(RMSProp)optimizer to improve the training efficiency of the detection model and enhance the classification performance of the detection model.The research results show that the intrusion detection accuracy of the research model is 96.41%-97.18%.It can be seen that after optimizing the research model,intrusion detection technology has significantly improved both the efficiency and accuracy of model training.The research model can classify data based on access sources,which improves the classifi-cation performance of intrusion detection models for access behavior.At the same time,the improvement of classification performance optimizes the recognition efficiency of computers for attack behavior,enhances the defense effect of computers,and effectively guaran-tees the network security environment of users.Therefore,the research provides a technical method with good recognition effect for the detection of network intrusion behavior.
作者
范申民
王磊
张芬
FAN Shenmin;WANG Lei;ZHANG Fen(Shangluo University,Shangluo Shanxi 726000,China)
出处
《自动化与仪器仪表》
2025年第5期7-11,共5页
Automation & Instrumentation
基金
陕西省商洛市科技局科研项目《基于层次分析的秦岭康养指数的编制方法及研究(以商洛为例)》(2022-Y-0015)
教育部产学合作协同育人项目《学生学习诉求与信息技术嵌入式建构下高校瑜伽》(230907307124852)。
