Data Mining and Knowledge Discovery

, Volume 30, Issue 4, pp 891–927

On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study

  • Guilherme O. Campos
  • Arthur Zimek
  • Jörg Sander
  • Ricardo J. G. B. Campello
  • Barbora Micenková
  • Erich Schubert
  • Ira Assent
  • Michael E. Houle
Article

DOI: 10.1007/s10618-015-0444-8

Cite this article as:
Campos, G.O., Zimek, A., Sander, J. et al. Data Min Knowl Disc (2016) 30: 891. doi:10.1007/s10618-015-0444-8

Abstract

The evaluation of unsupervised outlier detection algorithms is a constant challenge in data mining research. Little is known regarding the strengths and weaknesses of different standard outlier detection models, and the impact of parameter choices for these algorithms. The scarcity of appropriate benchmark datasets with ground truth annotation is a significant impediment to the evaluation of outlier methods. Even when labeled datasets are available, their suitability for the outlier detection task is typically unknown. Furthermore, the biases of commonly-used evaluation measures are not fully understood. It is thus difficult to ascertain the extent to which newly-proposed outlier detection methods improve over established methods. In this paper, we perform an extensive experimental study on the performance of a representative set of standard k nearest neighborhood-based methods for unsupervised outlier detection, across a wide variety of datasets prepared for this purpose. Based on the overall performance of the outlier detection methods, we provide a characterization of the datasets themselves, and discuss their suitability as outlier detection benchmark sets. We also examine the most commonly-used measures for comparing the performance of different methods, and suggest adaptations that are more suitable for the evaluation of outlier detection results.

Keywords

Unsupervised outlier detection Evaluation Measures Datasets 

Copyright information

© The Author(s) 2016

Authors and Affiliations

  • Guilherme O. Campos
    • 1
  • Arthur Zimek
    • 2
  • Jörg Sander
    • 3
  • Ricardo J. G. B. Campello
    • 1
  • Barbora Micenková
    • 4
  • Erich Schubert
    • 2
  • Ira Assent
    • 4
  • Michael E. Houle
    • 5
  1. 1.University of São PauloSão CarlosBrazil
  2. 2.Ludwig-Maximilians-Universität MünchenMunichGermany
  3. 3.Department of Computing ScienceUniversity of AlbertaEdmontonCanada
  4. 4.Department of Computer ScienceAarhus UniversityAarhusDenmark
  5. 5.National Institute of InformaticsChiyoda-kuJapan

Personalised recommendations