一种新型的自适应入侵检测系统的研制

针对当前入侵监测系统存在的自适应能力较差、扩展性差的问题，基于面向混合类型数据的快速启发式聚类算法FHCAM和属性约束的模糊规则挖掘算法ACFMAR，提出了一种采用数据挖掘技术的自适应入侵检测系统DMAIDS．该系统通过划分聚类的方法划分出异常入侵记录；模糊关联规则的方法提取入侵模式．通过对1999年举行的数据挖掘大赛所使用的10%子集进行实验，结果表明该系统平均检测率和平均误检率比大赛获得冠军检测方法准确率提高了近2倍，检测率从数据子集1的65．25％自适应提高到数据子集9的85．7％能自适应的检测各种攻击，表明该系统具有很好的应用前景。

Realization of a Novel Adaptive Intrusion Detection System

Based the data mining technology, This paper adapt a novel IDS system that make use of heuristic clustering Algorithm and fuzzy mining algorithm. Verification tests are carried out by using the 10% subset of KDD Cup 1999 Data Set, The average detection rate is 71.67% and the average false detection rate is 0.92%.It is about twice as precise as the best system in the KDD Cup. And the detection rate increases from 65.25% (the first subset) to 85.7% (the ninth subset) adaptively. The experimental results reveal that the system can detect variety of intrusions quickly and has a bright future to be used in different areas.