Internet Traffic Classification Using DBSCAN
Authors:
- Caihong YangCYCaihong Yang
- Fei WangFWFei Wang
- Benxiong HuangBHBenxiong Huang
- 1 July 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 163-166
Abstract
In recent years, a technique based on machine learning for Internet traffic classification has attracted more and more attentions. It not only overcomes some short comings of traditional classification technique based on port number,but also does not inspect the packet payload, which involves the security and privacy. In this paper, we apply an unsupervised machine learning approach based on DBSCAN algorithm. DBSCAN algorithm has three merits: (1) minimal requirements of domain knowledge to determine the input parameters; (2) discovery of clusters with arbitrary shapes; (3)good efficiency on large data set. Experiment results show that DBSCAN has better effectiveness and efficiency.Keywords
This publication has 11 references indexed in Scilit:
- Identifying and discriminating between web and peer-to-peer traffic in the network core2007
- QRP05-4: Internet Traffic Identification using Machine LearningIEEE Globecom 2006, 2006
- Evaluation of an Efficient Measurement Concept for P2P Multiprotocol Traffic Analysis2006
- Traffic classification using clustering algorithms2006
- Identifying Known and Unknown Peer-to-Peer Traffic2006
- Advanced P2P Multiprotocol Traffic Analysis Based on Application Level Signature Detection2006
- ACAS2005
- Automated traffic classification and application identification using machine learning2005
- Transport layer identification of P2P traffic2004
- Accurate, scalable in-network identification of p2p traffic using application signatures2004