文章基本信息

标题：Impact of Asymmetry of Internet Traffic for Heuristic Based Classification
本地全文：下载
作者：Chris Richter ; Michael Finsterbusch ; Klaus Hänßgen 等
期刊名称：International Journal of Computer Networks (IJCN)
电子版ISSN：1985-4129
出版年度：2012
卷号：4
期号：5
页码：167-176
出版社：Computer Science Journals
摘要：Accurate traffic classification is necessary for many administrative networking tasks like security monitoring, providing Quality of Service and network design or planning. In this paper we illustrate the accuracy of 18 different machine learning algorithms with different statistical parameter combinations. Additionally, we divide the statistical parameters into upstream and downstream to observe the influence of the protocol inherent differences of client and server behaviour for traffic classification. Our results show that this differentiation can increase the protocol detection rate and decrement the processing time.
关键词：Flow Classification; Internet Traffic; Traffic Identification