Visual evaluation of outlier detection models

Elke Achtert*, Hans Peter Kriegel, Lisa Reichert, Erich Schubert, Remigius Wojdanowski, Arthur Zimek

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Many outlier detection methods do not merely provide the decision for a single data object being or not being an outlier. Instead, many approaches give an "outlier score" or "outlier factor" indicating "how much" the respective data object is an outlier. Such outlier scores differ widely in their range, contrast, and expressiveness between different outlier models. Even for one and the same outlier model, the same score can indicate a different degree of "outlierness" in different data sets or regions of different characteristics in one data set. Here, we demonstrate a visualization tool based on a unification of outlier scores that allows to compare and evaluate outlier scores visually even for high dimensional data.

Original languageEnglish
Title of host publicationDatabase Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings
EditorsH. Kitagawa, Y. Ishikawa, Q. Li, C. Watanabe
PublisherSpringer
Publication date28. Dec 2010
EditionPART 2
Pages396-399
ISBN (Print)978-3-642-12097-8
ISBN (Electronic)978-3-642-12098-5
DOIs
Publication statusPublished - 28. Dec 2010
Externally publishedYes
Event15th International Conference on Database Systems for Advanced Applications - Tsukuba, Japan
Duration: 1. Apr 20104. Apr 2010

Conference

Conference15th International Conference on Database Systems for Advanced Applications
CountryJapan
CityTsukuba
Period01/04/201004/04/2010
SponsorKIISE Database Society of Korea, China Computer Federation Database Technical Committee, ARC Research Network in Enterprise Information Infrastructure, Asian Institute of Technology (AIT), Information Processing Society of Japan (IPSJ)
SeriesLecture Notes in Computer Science
Volume5982
ISSN0302-9743

Fingerprint

Visualization

Cite this

Achtert, E., Kriegel, H. P., Reichert, L., Schubert, E., Wojdanowski, R., & Zimek, A. (2010). Visual evaluation of outlier detection models. In H. Kitagawa, Y. Ishikawa, Q. Li, & C. Watanabe (Eds.), Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings (PART 2 ed., pp. 396-399). Springer. Lecture Notes in Computer Science, Vol.. 5982 https://doi.org/10.1007/978-3-642-12098-5_34
Achtert, Elke ; Kriegel, Hans Peter ; Reichert, Lisa ; Schubert, Erich ; Wojdanowski, Remigius ; Zimek, Arthur. / Visual evaluation of outlier detection models. Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings. editor / H. Kitagawa ; Y. Ishikawa ; Q. Li ; C. Watanabe. PART 2. ed. Springer, 2010. pp. 396-399 (Lecture Notes in Computer Science, Vol. 5982).
@inproceedings{87b09d69871c44b68f935c5897ce107d,
title = "Visual evaluation of outlier detection models",
abstract = "Many outlier detection methods do not merely provide the decision for a single data object being or not being an outlier. Instead, many approaches give an {"}outlier score{"} or {"}outlier factor{"} indicating {"}how much{"} the respective data object is an outlier. Such outlier scores differ widely in their range, contrast, and expressiveness between different outlier models. Even for one and the same outlier model, the same score can indicate a different degree of {"}outlierness{"} in different data sets or regions of different characteristics in one data set. Here, we demonstrate a visualization tool based on a unification of outlier scores that allows to compare and evaluate outlier scores visually even for high dimensional data.",
author = "Elke Achtert and Kriegel, {Hans Peter} and Lisa Reichert and Erich Schubert and Remigius Wojdanowski and Arthur Zimek",
year = "2010",
month = "12",
day = "28",
doi = "10.1007/978-3-642-12098-5_34",
language = "English",
isbn = "978-3-642-12097-8",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "396--399",
editor = "H. Kitagawa and Y. Ishikawa and Q. Li and C. Watanabe",
booktitle = "Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings",
address = "Germany",
edition = "PART 2",

}

Achtert, E, Kriegel, HP, Reichert, L, Schubert, E, Wojdanowski, R & Zimek, A 2010, Visual evaluation of outlier detection models. in H Kitagawa, Y Ishikawa, Q Li & C Watanabe (eds), Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings. PART 2 edn, Springer, Lecture Notes in Computer Science, vol. 5982, pp. 396-399, 15th International Conference on Database Systems for Advanced Applications, Tsukuba, Japan, 01/04/2010. https://doi.org/10.1007/978-3-642-12098-5_34

Visual evaluation of outlier detection models. / Achtert, Elke; Kriegel, Hans Peter; Reichert, Lisa; Schubert, Erich; Wojdanowski, Remigius; Zimek, Arthur.

Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings. ed. / H. Kitagawa; Y. Ishikawa; Q. Li; C. Watanabe. PART 2. ed. Springer, 2010. p. 396-399 (Lecture Notes in Computer Science, Vol. 5982).

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Visual evaluation of outlier detection models

AU - Achtert, Elke

AU - Kriegel, Hans Peter

AU - Reichert, Lisa

AU - Schubert, Erich

AU - Wojdanowski, Remigius

AU - Zimek, Arthur

PY - 2010/12/28

Y1 - 2010/12/28

N2 - Many outlier detection methods do not merely provide the decision for a single data object being or not being an outlier. Instead, many approaches give an "outlier score" or "outlier factor" indicating "how much" the respective data object is an outlier. Such outlier scores differ widely in their range, contrast, and expressiveness between different outlier models. Even for one and the same outlier model, the same score can indicate a different degree of "outlierness" in different data sets or regions of different characteristics in one data set. Here, we demonstrate a visualization tool based on a unification of outlier scores that allows to compare and evaluate outlier scores visually even for high dimensional data.

AB - Many outlier detection methods do not merely provide the decision for a single data object being or not being an outlier. Instead, many approaches give an "outlier score" or "outlier factor" indicating "how much" the respective data object is an outlier. Such outlier scores differ widely in their range, contrast, and expressiveness between different outlier models. Even for one and the same outlier model, the same score can indicate a different degree of "outlierness" in different data sets or regions of different characteristics in one data set. Here, we demonstrate a visualization tool based on a unification of outlier scores that allows to compare and evaluate outlier scores visually even for high dimensional data.

U2 - 10.1007/978-3-642-12098-5_34

DO - 10.1007/978-3-642-12098-5_34

M3 - Article in proceedings

AN - SCOPUS:78650492208

SN - 978-3-642-12097-8

T3 - Lecture Notes in Computer Science

SP - 396

EP - 399

BT - Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings

A2 - Kitagawa, H.

A2 - Ishikawa, Y.

A2 - Li, Q.

A2 - Watanabe, C.

PB - Springer

ER -

Achtert E, Kriegel HP, Reichert L, Schubert E, Wojdanowski R, Zimek A. Visual evaluation of outlier detection models. In Kitagawa H, Ishikawa Y, Li Q, Watanabe C, editors, Database Systems for Advanced Applications - 15th International Conference, DASFAA 2010, Proceedings. PART 2 ed. Springer. 2010. p. 396-399. (Lecture Notes in Computer Science, Vol. 5982). https://doi.org/10.1007/978-3-642-12098-5_34