基于Agent的网络入侵检测专家系统

Intrusion Detection Expert System of Network Based on Agent

摘要: 大多数网络入侵检测专家系统采用单一的推理机制.而基于规则推理和基于事例推理这两种推理机制各有优缺点, 采用单一推理机制会降低入侵检测的准确性.文中提出了基于规则和事例的入侵检测混合推理机制.在此基础上, 设计了一个基于Agent的网络入侵检测专家系统 (AIDES).实验结果表明, 文中设计的AIDES是有效的.

Abstract: Most intrusion detection expert system of network adopted single reasoning mechanism.However, rule-based reasoning or case-based reasoning had advantages and disadvantages.The adoption of single reasoning mechanism would lead to inaccurate intrusion detection.The paper proposed mixed reasoning mechanism based on rule-based reasoning and case-based reasoning.On the base of this, the paper designed an intrusion detection expert system of network based on agent, namely AIDES.The experiments indicate that AIDES is effective.