关键词: 人工免疫算法 入侵检测 动态克隆选择

Abstract:

N-AIS can he only used as a misuse detector to detect the given static antigen data set, hut cannot automatically adapt to the dynamic network intrusion. The paper introduces the dynamic elonal selection algorithm called DynamiCS in order to extend the N-AIS. The effects of three important system para meters of toleration period, activation threshold and life span are tested and analyzed on artificially created scenarios that reflect two common real c ircumstances of IDS. The abilities for DynamiCS to perform incremental learning on converged data and to adapt to novel data are demonstrated.

Key words: artificial immune algorithm, intrusion detection, dynamic clonal selection