QQ客服 官方微博 反馈留言

	Efficient attack detection based on a compressed model
	Jin, Shichao ; Kim, Okhee ; Chen, Tieming
	2013
英文摘要	In order to achieve the goal of high efficiency in intrusion detection systems, especially in the real-time attack detection environment, a compressed model is proposed in this paper. With the emergence of the new clustering methods, such as the affinity propagation, the idea of the compressed detection model tends to be mature as it is unnecessary to define the number of centers beforehand. The compressed model resulting from both the horizontal compression and the vertical compression is built with representative training data and useful attributes in each package. In addition, a distance matrix is extracted from previous steps for processing complex data. Experimental study based on two publicly available datasets presents that the compressed model proposed can effectively speed up the detection procedure (up to 184 times) and most importantly, a minimal accuracy difference is guaranteed as well (less than 1% on average). ? 2013 Springer-Verlag.; EI; 0
语种	英语
DOI标识	10.1007/978-3-642-38033-4_18
内容类型	其他
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/325787]
专题	软件与微电子学院
推荐引用方式 GB/T 7714	Jin, Shichao,Kim, Okhee,Chen, Tieming. Efficient attack detection based on a compressed model. 2013-01-01.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。