摘要： 针对现有入侵检测方法的缺陷，结合异常检测和误用检测，提出了一种用数据挖掘技术构造入侵检测系统的方法，使用该方法构造了一个基于数据挖掘的入侵检测原型系统。实验表明，该系统对已知攻击模式具有很高的检测率，对未知攻击模式也具有一定的检测能力。同时，该系统也具有一定的智能性和自适应性。

关键词: 网络安全, 入侵检测, 数据挖掘, 贝叶斯分类

Abstract: Allowing for shortcomings of existing intrusion detection approaches, this paper combines anomaly detection with misuse detection. An approach which uses data mining technology to build IDS is presented and a prototype is given. Experiment shows that the prototype can detect known intrusion modes with very high detection rate, and also can detect some unknown intrusion modes. At the same time, the system is more intelligent and self-adaptable.

中图分类号: 

• TP393.08