微型电脑应用

传统的网络入侵行为检测方法检测时间长、检测率低,导致网络安全受到严重威胁,因此基于贝叶斯分类算法对网络入侵行为检测方法进行研究。首先设计用户行为日志采集系统,通过采集节点模块、数据分析与存储模块对用户行为日志数据进行采集,其次通过消除冗余数据、数据规范化等环节预处理采集到的数据。针对预处理后的数据,构建基于朴素贝叶斯分类的网络入侵行为检测模型,基于不同的属性集构建非网络入侵与网络入侵分类规则,实现网络入侵行为检测。实验结果显示该方法检测结果准确度在97%以上,检测过程花费时间与对比方法相比降低4 s以上。

The traditional network intrusion detection method has a long detection time and low detection rate, which leads to a serious threat to the network security. This paper studies the network intrusion detection method based on Bayesian classification algorithm. Firstly, the user behavior log collection system is designed, itcollects the user behavior log data through the collection node module, data analysis and storage module, and then preprocesses the collected data by eliminating redundant data and data normalization.According to the preprocessed data, the network intrusion detection model based on naive Bayesian classification is constructed, and the non network intrusion and network intrusion classification rules based on different attribute sets are constructed to realize the network intrusion detection.The experimental results show that the accuracy of the method is more than 97%, and the detection process takes more than 4 s less time than the comparison method.