Application of data mining for the analysis of Internet path performance

Author

Borzemski, Leszek ; Lubczynski, Lukasz ; Nowak, Ziemowit

Author_Institution

Inst. of Control & Syst. Eng., Wroclaw Univ. of Technol., Poland

fYear

2004

fDate

11-13 Feb. 2004

Firstpage

54

Lastpage

59

Abstract

We apply data mining analysis to study Internet path performance. We show how a data mining system can be used by end-users in this application. The traceroute packet probing technique is used for Internet measurements. The data set is mined using neural clustering and tree classification mining functions available in IBM Intelligent Miner. We discover from the measured data sets how the round-trip times of the packets and the number of hops they pass vary with the day of the week and the time of the measurement. The decision tree model shows good accuracy of 97% and may be useful to predict Internet path performance.

Keywords

Internet; data mining; performance evaluation; IBM Intelligent Miner; Internet path performance; data mining; decision tree classification mining; end-users; neural clustering; traceroute packet probing technique; Control systems; Data analysis; Data engineering; Data mining; Delay; IP networks; Internet; Performance analysis; Performance evaluation; Time measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel, Distributed and Network-Based Processing, 2004. Proceedings. 12th Euromicro Conference on

ISSN

1066-6192

Print_ISBN

0-7695-2083-9

Type

conf

DOI

10.1109/EMPDP.2004.1271427

Filename

1271427