Presentation 2009-07-17
Visualizing and Classifying Internet Traffic based on Connection Patterns
Yosuke HIMURA, Kensuke FUKUDA, Kenjiro CHO, Hiroshi ESAKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this technical report, we develop a visualization tool of Internet traffic based on connection patterns for additional information of anomaly detection reports. The connection pattern is defined as information aggregated from flows generated by a certain host; it represents how the 5-tuples are correlated each other. We evaluate our tool by using anomaly detection algorithm based on Sketch and the multi-scale gamma model with MAWI real-traffic traces. The main result of the visualization and investigation of anomaly traffic is as follows: (a) efficient profiling of anomaly traffic, (b) identification of anomaly traffic by using visualized connection patterns, (c) support for identifing unknown traffic and finding misclassified traffic. We aim to construct a database of anomaly traffic, and to automatically tune parameters of network anomaly detection and classification methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) traffic analysis / network anomaly detection / traffic classification
Paper # IA2009-27
Date of Issue

Conference Information
Committee IA
Conference Date 2009/7/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Internet Architecture(IA)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Visualizing and Classifying Internet Traffic based on Connection Patterns
Sub Title (in English)
Keyword(1) traffic analysis
Keyword(2) network anomaly detection
Keyword(3) traffic classification
1st Author's Name Yosuke HIMURA
1st Author's Affiliation The University of Tokyo()
2nd Author's Name Kensuke FUKUDA
2nd Author's Affiliation National Institute of Informatics:PRESTO, JST
3rd Author's Name Kenjiro CHO
3rd Author's Affiliation Internet Initiative Japan
4th Author's Name Hiroshi ESAKI
4th Author's Affiliation The University of Tokyo
Date 2009-07-17
Paper # IA2009-27
Volume (vol) vol.109
Number (no) 137
Page pp.pp.-
#Pages 6
Date of Issue