A comparative evaluation of outlier detection algorithms: Experiments and analyses

Rémi Domingues; Maurizio Filippone; Pietro Michiardi; Jihane Zouaoui

doi:10.1016/j.patcog.2017.09.037

A comparative evaluation of outlier detection algorithms: Experiments and analyses

Rémi Domingues^*, Maurizio Filippone, Pietro Michiardi, Jihane Zouaoui

^*Corresponding author for this work

Statistics

Research output: Contribution to journal › Article › peer-review

364 Scopus citations

Abstract

We survey unsupervised machine learning algorithms in the context of outlier detection. This task challenges state-of-the-art methods from a variety of research fields to applications including fraud detection, intrusion detection, medical diagnoses and data cleaning. The selected methods are benchmarked on publicly available datasets and novel industrial datasets. Each method is then submitted to extensive scalability, memory consumption and robustness tests in order to build a full overview of the algorithms’ characteristics.

Original language	English (US)
Pages (from-to)	406-421
Number of pages	16
Journal	Pattern Recognition
Volume	74
DOIs	https://doi.org/10.1016/j.patcog.2017.09.037
State	Published - Feb 2018

Bibliographical note

Publisher Copyright:
© 2017 Elsevier Ltd

Keywords

Fraud detection
Isolation forest
Novelty detection
Outlier detection
Variational inference

ASJC Scopus subject areas

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.1016/j.patcog.2017.09.037

Cite this

@article{55dadde9889f4fd6a3eea9fc1714f327,

title = "A comparative evaluation of outlier detection algorithms: Experiments and analyses",

abstract = "We survey unsupervised machine learning algorithms in the context of outlier detection. This task challenges state-of-the-art methods from a variety of research fields to applications including fraud detection, intrusion detection, medical diagnoses and data cleaning. The selected methods are benchmarked on publicly available datasets and novel industrial datasets. Each method is then submitted to extensive scalability, memory consumption and robustness tests in order to build a full overview of the algorithms{\textquoteright} characteristics.",

keywords = "Fraud detection, Isolation forest, Novelty detection, Outlier detection, Variational inference",

author = "R{\'e}mi Domingues and Maurizio Filippone and Pietro Michiardi and Jihane Zouaoui",

note = "Publisher Copyright: {\textcopyright} 2017 Elsevier Ltd",

year = "2018",

month = feb,

doi = "10.1016/j.patcog.2017.09.037",

language = "English (US)",

volume = "74",

pages = "406--421",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - A comparative evaluation of outlier detection algorithms

T2 - Experiments and analyses

AU - Domingues, Rémi

AU - Filippone, Maurizio

AU - Michiardi, Pietro

AU - Zouaoui, Jihane

PY - 2018/2

Y1 - 2018/2

N2 - We survey unsupervised machine learning algorithms in the context of outlier detection. This task challenges state-of-the-art methods from a variety of research fields to applications including fraud detection, intrusion detection, medical diagnoses and data cleaning. The selected methods are benchmarked on publicly available datasets and novel industrial datasets. Each method is then submitted to extensive scalability, memory consumption and robustness tests in order to build a full overview of the algorithms’ characteristics.

AB - We survey unsupervised machine learning algorithms in the context of outlier detection. This task challenges state-of-the-art methods from a variety of research fields to applications including fraud detection, intrusion detection, medical diagnoses and data cleaning. The selected methods are benchmarked on publicly available datasets and novel industrial datasets. Each method is then submitted to extensive scalability, memory consumption and robustness tests in order to build a full overview of the algorithms’ characteristics.

KW - Fraud detection

KW - Isolation forest

KW - Novelty detection

KW - Outlier detection

KW - Variational inference

UR - http://www.scopus.com/inward/record.url?scp=85032291155&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2017.09.037

DO - 10.1016/j.patcog.2017.09.037

M3 - Article

AN - SCOPUS:85032291155

SN - 0031-3203

VL - 74

SP - 406

EP - 421

JO - Pattern Recognition

JF - Pattern Recognition

ER -

A comparative evaluation of outlier detection algorithms: Experiments and analyses

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this