Analyzing statistical effect of sampling on network traffic dataset

Raman Singh*, Harish Kumar, R. K. Singla

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)

Abstract

In sampling of huge network traffic dataset, some packets are chosen out of total packets. Leftover packets may have effect on statistical characteristics of the data. In this paper effect of sampling on statistical characteristics is discussed. A well-known benchmarked NSL KDD network traffic dataset is used. Three sampling techniques namely - random, systematic and under-over sampling are used. Various attributes of dataset considered are duration, src_bytes, dst_bytes, wrong_fragment, num_compromised, num_file_ creations and srv_count. Parameter of statistical characteristics like range, mean and standard deviation is used for analysis purpose. Result shows that sampling has considerable statistical effect on network traffic dataset.

Original languageEnglish
Title of host publicationICT and Critical Infrastructure
Subtitle of host publicationProceedings of the 48th Annual Convention of Computer Society of India
EditorsS. Satapathy, S. Avadhani, P. Udgata, S. Lakshminarayana
PublisherSpringer-Verlag
Pages401-408
Number of pages8
Volume1
ISBN (Electronic)9783319031071
ISBN (Print)9783319031064
DOIs
Publication statusPublished - 2014
Externally publishedYes
Event48th Annual Convention of Computer Society of India, CSI 2013 - Visakhapatnam, India
Duration: 13 Dec 201315 Dec 2013

Publication series

NameAdvances in Intelligent Systems and Computing
PublisherSpringer
Volume248
ISSN (Print)2194-5357

Conference

Conference48th Annual Convention of Computer Society of India, CSI 2013
Country/TerritoryIndia
CityVisakhapatnam
Period13/12/1315/12/13

Keywords

  • intrusion detection system
  • network traffic dataset
  • sampling

Fingerprint

Dive into the research topics of 'Analyzing statistical effect of sampling on network traffic dataset'. Together they form a unique fingerprint.

Cite this