An Experimental Research of Traffic Identification Algorithms in Broadband Network

Author

Chen, Na ; Shou, Guochu ; Hu, Yihong ; Guo, Zhigang

Author_Institution

Sch. of Inf. & Commun. Eng., Beijing Univ. of Posts & Telecommun. (BUPT), Beijing, China

fYear

2009

fDate

18-20 Jan. 2009

Firstpage

1

Lastpage

4

Abstract

Adopting the machine learning method to identify the traffic in the broadband network has become a new research direction in this field. In this paper, we evaluate 11 kinds of supervised learning algorithm of the machine learning through the index of testing time, modeling time, accuracy rate and CPU utilization ratio, then select suitable algorithms for the broadband network. Through the analysis we draw the conclusion that the algorithms of C4.5, RandomTree, OneR and BayesNet fit the traffic identification in the broadband network.

Keywords

Internet; broadband networks; learning (artificial intelligence); telecommunication traffic; BayesNet algorithm; C4.5 algorithm; CPU utilization ratio; OneR algorithm; RandomTree algorithm; accuracy rate; broadband network; experimental research; machine learning method; modeling time; supervised learning algorithm; testing time index; traffic identification algorithms; Broadband communication; Classification tree analysis; Decision trees; Learning systems; Machine learning; Machine learning algorithms; Supervised learning; Telecommunication traffic; Testing; Traffic control;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Network and Multimedia Technology, 2009. CNMT 2009. International Symposium on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-5272-9

Type

conf

DOI

10.1109/CNMT.2009.5374758

Filename

5374758