作者: Alberto Dainotti , Walter de Donato , Antonio Pescapé
DOI: 10.1007/978-3-642-01645-5_8
关键词: Computer science 、 Traffic classification 、 Plug-in 、 Identification (information) 、 Hash table 、 Implementation 、 Software 、 Data science 、 Port (computer networking) 、 Architecture 、 Data mining
摘要: The research on network traffic classification has recently become very active. community, moved by increasing difficulties in the automated identification of traffic, started to investigate approaches alternative port-based and payload-based techniques. Despite large quantity works published past few years this topic, implementations targeting have been made available community. Moreover, most proposed literature suffer problems related ability evaluating comparing them. In paper we present a novel community-oriented software for called TIE, which aims at becoming common tool fair evaluation comparison different techniques fostering sharing data. TIE supports combination more plugins order build multi-classifier systems, its architecture is designed allow online classification.