On the Correlation Between Local Intrinsic Dimensionality and Outlierness

Michael E. Houle, Erich Schubert, Arthur Zimek

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

Abstract

Data mining methods for outlier detection are usually based on non-parametric density estimates in various variations. Here we argue for the use of local intrinsic dimensionality as a measure of outlierness and demonstrate empirically that it is a meaningful alternative and complement to classic methods.

Original languageEnglish
Title of host publicationSimilarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings
EditorsStéphane Marchand-Maillet, Yasin N. Silva, Edgar Chávez
PublisherSpringer VS
Publication date2018
Pages177-191
ISBN (Print)9783030022235
DOIs
Publication statusPublished - 2018
Event11th International Conference on Similarity Search and Applications, SISAP 2018 - Lima, Peru
Duration: 7 Oct 20189 Oct 2018

Conference

Conference11th International Conference on Similarity Search and Applications, SISAP 2018
CountryPeru
CityLima
Period07/10/201809/10/2018
SeriesLecture Notes in Computer Science
Volume11223
ISSN0302-9743

Fingerprint

data mining
outlier
method
detection

Keywords

  • Comparison
  • Intrinsic dimensionality
  • Outlier detection

Cite this

Houle, M. E., Schubert, E., & Zimek, A. (2018). On the Correlation Between Local Intrinsic Dimensionality and Outlierness. In S. Marchand-Maillet, Y. N. Silva, & E. Chávez (Eds.), Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings (pp. 177-191). Springer VS. Lecture Notes in Computer Science, Vol.. 11223 https://doi.org/10.1007/978-3-030-02224-2_14
Houle, Michael E. ; Schubert, Erich ; Zimek, Arthur. / On the Correlation Between Local Intrinsic Dimensionality and Outlierness. Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings. editor / Stéphane Marchand-Maillet ; Yasin N. Silva ; Edgar Chávez. Springer VS, 2018. pp. 177-191 (Lecture Notes in Computer Science, Vol. 11223).
@inproceedings{3ea900d5b91e45d49bc3e081ff4cb734,
title = "On the Correlation Between Local Intrinsic Dimensionality and Outlierness",
abstract = "Data mining methods for outlier detection are usually based on non-parametric density estimates in various variations. Here we argue for the use of local intrinsic dimensionality as a measure of outlierness and demonstrate empirically that it is a meaningful alternative and complement to classic methods.",
keywords = "Comparison, Intrinsic dimensionality, Outlier detection",
author = "Houle, {Michael E.} and Erich Schubert and Arthur Zimek",
year = "2018",
doi = "10.1007/978-3-030-02224-2_14",
language = "English",
isbn = "9783030022235",
pages = "177--191",
editor = "St{\'e}phane Marchand-Maillet and Silva, {Yasin N.} and Edgar Ch{\'a}vez",
booktitle = "Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings",
publisher = "Springer VS",

}

Houle, ME, Schubert, E & Zimek, A 2018, On the Correlation Between Local Intrinsic Dimensionality and Outlierness. in S Marchand-Maillet, YN Silva & E Chávez (eds), Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings. Springer VS, Lecture Notes in Computer Science, vol. 11223, pp. 177-191, 11th International Conference on Similarity Search and Applications, SISAP 2018, Lima, Peru, 07/10/2018. https://doi.org/10.1007/978-3-030-02224-2_14

On the Correlation Between Local Intrinsic Dimensionality and Outlierness. / Houle, Michael E.; Schubert, Erich; Zimek, Arthur.

Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings. ed. / Stéphane Marchand-Maillet; Yasin N. Silva; Edgar Chávez. Springer VS, 2018. p. 177-191.

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

TY - GEN

T1 - On the Correlation Between Local Intrinsic Dimensionality and Outlierness

AU - Houle, Michael E.

AU - Schubert, Erich

AU - Zimek, Arthur

PY - 2018

Y1 - 2018

N2 - Data mining methods for outlier detection are usually based on non-parametric density estimates in various variations. Here we argue for the use of local intrinsic dimensionality as a measure of outlierness and demonstrate empirically that it is a meaningful alternative and complement to classic methods.

AB - Data mining methods for outlier detection are usually based on non-parametric density estimates in various variations. Here we argue for the use of local intrinsic dimensionality as a measure of outlierness and demonstrate empirically that it is a meaningful alternative and complement to classic methods.

KW - Comparison

KW - Intrinsic dimensionality

KW - Outlier detection

U2 - 10.1007/978-3-030-02224-2_14

DO - 10.1007/978-3-030-02224-2_14

M3 - Article in proceedings

SN - 9783030022235

SP - 177

EP - 191

BT - Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings

A2 - Marchand-Maillet, Stéphane

A2 - Silva, Yasin N.

A2 - Chávez, Edgar

PB - Springer VS

ER -

Houle ME, Schubert E, Zimek A. On the Correlation Between Local Intrinsic Dimensionality and Outlierness. In Marchand-Maillet S, Silva YN, Chávez E, editors, Similarity Search and Applications - 11th International Conference, SISAP 2018, Proceedings. Springer VS. 2018. p. 177-191. (Lecture Notes in Computer Science, Vol. 11223). https://doi.org/10.1007/978-3-030-02224-2_14