摘要
无法检测到未知攻击以及不能自动更新知识库是现有误用检测系统的两大缺点 .概念级误用检测系统(CLMDS)中利用SRRW特征选取算法、CHGL技术和独立双模型互训练结构极大地提升了系统的认知能力 ,有效地解决了上述问题 .文章从静态和动态两个层面对系统的认知能力进行了分析 ;实验结果表明 :CLMDS具有很强的认知能力 ,不但能检测到未知的攻击样式 ,而且还能实现知识库的自动更新 .
Current misuse detection systems are of little use for new attacks and they cannot automatically update their rule databases.SRRW,CHGL and the technology of co-training for independent dual-model greatly improve the apperception ability of CLMDS,and provide a good solution for the limitation of misuse detection systems.Apperception ability is analyzed from static aspect as well as dynamic aspect.Results of the experiments show that with powerful apperception ability CLMDS can not only detect new attacks but also update its own rule base automatically.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2004年第10期1694-1697,共4页
Acta Electronica Sinica
基金
普天首信重大科研项目 (No.0 2 1 1 2 5)