Visualization of anomaly detection using prediction sensitivity

Pavel Laskov, Konrad Rieck, Christin Schäfer, Klaus Robert Müller

Research output: Chapter in Book/Report/Conference proceedingConference contribution

23 Citations (Scopus)

Abstract

Visualization of learning-based intrusion detection methods is a challenging problem. In this paper we propose a novel method for visualization of anomaly detection and feature selection, based on prediction sensitivity. The method allows an expert to discover informative features for separation of normal and attack instances. Experiments performed on the KDD Cup dataset show that explanations provided by prediction sensitivity reveal the nature of attacks. Application of prediction sensitivity for feature selection yields a major improvement of detection accuracy.

Original languageEnglish
Title of host publicationSICHERHEIT 2005 - Sicherheit - Schutz und Zuverlassigkeit, Beitrage der 2. Jahrestagung des Fachbereichs Sicherheit der Gesellschaft fur Informatik e.V. (GI)
Pages197-208
Number of pages12
Publication statusPublished - 2005
EventSICHERHEIT 2005 - Sicherheit - Schutz und Zuverlassigkeit, Beitrage der 2. Jahrestagung des Fachbereichs Sicherheit der Gesellschaft fur Informatik e.V. (GI)- 2nd Annual Meeting of the Department of Security of the Society for Informatics - Security - Regensburg, Germany
Duration: 2005 Apr 52005 Apr 8

Publication series

NameLecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)
ISSN (Print)1617-5468

Other

OtherSICHERHEIT 2005 - Sicherheit - Schutz und Zuverlassigkeit, Beitrage der 2. Jahrestagung des Fachbereichs Sicherheit der Gesellschaft fur Informatik e.V. (GI)- 2nd Annual Meeting of the Department of Security of the Society for Informatics - Security
CountryGermany
CityRegensburg
Period05/4/505/4/8

ASJC Scopus subject areas

  • Computer Science Applications

Fingerprint Dive into the research topics of 'Visualization of anomaly detection using prediction sensitivity'. Together they form a unique fingerprint.

  • Cite this

    Laskov, P., Rieck, K., Schäfer, C., & Müller, K. R. (2005). Visualization of anomaly detection using prediction sensitivity. In SICHERHEIT 2005 - Sicherheit - Schutz und Zuverlassigkeit, Beitrage der 2. Jahrestagung des Fachbereichs Sicherheit der Gesellschaft fur Informatik e.V. (GI) (pp. 197-208). (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).