Adaptive sampling and learning for unsupervised outlier detection

Zhiruo Zhao, Chilukuri K Mohan, Kishan G. Mehrotra

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

2 Scopus citations

Abstract

Unsupervised outlier detection algorithms often suffer from high false positive detection rates. Ensemble approaches can be used to address this problem. This paper proposes a novel ensemble method which adopts the use of an adaptive sampling approach, and combines the outputs of individual anomaly detection algorithms by a weighted majority voting rule in a complete unsupervised context. Simulations on well-known benchmark problems show substantial improvement in performance.

Original languageEnglish (US)
Title of host publicationProceedings of the 29th International Florida Artificial Intelligence Research Society Conference, FLAIRS 2016
PublisherAAAI Press
Pages460-465
Number of pages6
ISBN (Electronic)9781577357568
StatePublished - 2016
Event29th International Florida Artificial Intelligence Research Society Conference, FLAIRS 2016 - Key Largo, United States
Duration: May 16 2016May 18 2016

Other

Other29th International Florida Artificial Intelligence Research Society Conference, FLAIRS 2016
CountryUnited States
CityKey Largo
Period5/16/165/18/16

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Adaptive sampling and learning for unsupervised outlier detection'. Together they form a unique fingerprint.

  • Cite this

    Zhao, Z., Mohan, C. K., & Mehrotra, K. G. (2016). Adaptive sampling and learning for unsupervised outlier detection. In Proceedings of the 29th International Florida Artificial Intelligence Research Society Conference, FLAIRS 2016 (pp. 460-465). AAAI Press.