摘要
针对传统入侵检测系统建模与更新需要大量人工参与,提出一种基于数据挖掘的无指导自适应入侵检测系统。系统通过有效结合聚类、关联规则数据挖掘方法,自动进行检测规则的提取。经实验表明,提出的方法具有较好的检测率、误报率。
An unsupervised and adaptive intrusion detection system based on data mining is proposed to solve the problem that the constructing and updating processes in traditional IDS highly depend on manual means. The detection rules are automatically extracted by combining clustering mining method with association rules mining method efficiently. The experiment results show that our method performs well on detection rate and false positive rate.
出处
《计算机应用与软件》
CSCD
2009年第4期253-256,共4页
Computer Applications and Software
关键词
异常检测
无指导学习
规则提取
数据挖掘
Anomaly detection Unsupervised learning Rule extraction Data mining
