基于强化学习的计算机网络恶意入侵行为检测方法

A method of Malicious Intrusion Detection in Computer Networks Based on Reinforcement Learning

传统网络入侵检测方法采用特征选择的方式获得入侵数据,由于缺乏对数据的降维处理,导致检测精度较低。因此,提出基于强化学习的计算机网络恶意入侵行为检测方法。首先,对数据进行归一化和标准化处理,构建自编码网络对数据进行降维;其次,提取网络数据特征,构建计算机网络恶意入侵行为检测模型;最后,进行实验对比分析。实验结果表明,采用提出的方法对入侵行为进行检测,检测误报率较低,具有较高的检测精度。

Traditional network intrusion detection methods use feature selection to obtain intrusion data.Due to the lack of dimensionality reduction,the detection accuracy is low.Therefore,a malicious intrusion detection method based on reinforcement learning is proposed.Firstly,the data is normalized and standardized,and a self coding network is constructed to reduce the dimension of the data.Secondly,the network data features are extracted to build a malicious intrusion detection model of computer network.Finally,the experiment is compared and analyzed.The experimental results show that the proposed method has low false alarm rate and high detection accuracy for intrusion detection.