入侵检测系统中的智能化方法

Intelligent Method in Intrusion Detection System

入侵检测系统很好地弥补了访问控制、身份认证等传统机制所不能解决的问题。目前的入侵检测技术正处在第一代技术向下一代技术的过渡时期,未来的入侵检测研究需要融合其他学科和技术领域的知识,充分利用许多成熟的信息智能处理技术。文中讨论和研究了三种典型的智能检测技术,其中对统计学方法、专家系统进行了总体概述,重点阐述数据挖掘技术中的关联规则分析、序列模式分析和数据分类分析的工作原理。

Intrusion detection system, as a supplement of traditional access control and identity authentication, provides critical protections from potential attempts to exploit computer resource vulnerabilities. Intrusion detection is going to the next generation and will combine methods from other fields, exploiting many mature information technology and artificial intelligence, such as statistics method, expert system, neural network, agent and data mining etc. in the future. In this paper, discuss three kinds of typical intellectual detection technologies. And statistics method, expert system are comparatively briefly introduced. The association rule analysis, sequence patterns analysis and data classification analysis of data mining technology are emphatically illustrated.