基于贝叶斯分类算法的网络入侵检测系统设计  

Design of Network Intrusion Detection System Based on Bayesian Classification Algorithm

作　　者：崔斌 CUI Bin(Nanyang Normal University,Nanyang Henan 473061,China)

机构地区：[1]南阳师范学院,河南南阳473061

出　　处：《信息与电脑》2022年第13期65-67,共3页Information & Computer

摘　　要：网络入侵检测是否高效的关键在于算法。稳定的算法能够通过较少的特征数据对链接进行精确识别,从而实现提前预警。传统网络入侵行为检测耗时长、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(Bayesian Classifier,BC)设计了入侵检测系统架构。该系统改进了传统检测方法,能够精简并融合大量多源数据,随后通过自适应调整阈值实现统一分析与管理。结果表明,与传统检测相比,该系统具有较强的扩展性与分布性,检测结果准确度提升,检测过程消耗时间显著降低。The key to the efficiency of network intrusion detection is the algorithm.The stable algorithm can accurately identify links through less characteristic data,so as to achieve early warning.Traditional network intrusion detection is time-consuming and low detection rate,which leads to serious threats to network security.Therefore,an intrusion detection system architecture based on Bayesian Classifier(BC)is designed.The system improves the traditional detection,can simplify and fuse a large number of multi-source data,and then realizes unified analysis and management by adaptively adjusting the threshold.The results show that the system has strong expansibility and distribution.Compared with the traditional detection,the accuracy of the system is improved,and the time consumption of the detection process is significantly reduced.

关 键 词：贝叶斯分类算法 网络入侵 特征选择 

分 类 号：TP393.2[自动化与计算机技术—计算机应用技术]