Quality of similarity rankings in time series

Thomas Bernecker*, Michael E. Houle, Hans Peter Kriegel, Peer Kröger, Matthias Renz, Erich Schubert, Arthur Zimek

*Corresponding author for this work

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

Abstract

Time series data objects can be interpreted as high- dimensional vectors, which allows the application of many traditional distance measures as well as more specialized measures. However, many distance functions are known to suffer from poor contrast in high-dimensional settings, putting their usefulness as similarity measures into question. On the other hand, shared-nearest-neighbor distances based on the ranking of data objects induced by some primary distance measure have been known to lead to improved performance in high-dimensional settings. In this paper, we study the performance of shared-neighbor similarity measures in the context of similarity search for time series data objects. Our findings are that the use of shared-neighbor similarity measures generally results in more stable performances than that of their associated primary distance measures.

Original languageEnglish
Title of host publicationAdvances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings
EditorsD. Pfoser
PublisherSpringer
Publication date19. Sep 2011
Pages422-440
ISBN (Print)978-3-642-22921-3
ISBN (Electronic)978-3-642-22922-0
DOIs
Publication statusPublished - 19. Sep 2011
Externally publishedYes
Event12th International Symposium on Advances in Spatial and Temporal Databases - Minneapolis, United States
Duration: 24. Aug 201126. Aug 2011

Conference

Conference12th International Symposium on Advances in Spatial and Temporal Databases
CountryUnited States
CityMinneapolis
Period24/08/201126/08/2011
SeriesLecture Notes in Computer Science
Volume6849
ISSN0302-9743

Fingerprint

Time series

Cite this

Bernecker, T., Houle, M. E., Kriegel, H. P., Kröger, P., Renz, M., Schubert, E., & Zimek, A. (2011). Quality of similarity rankings in time series. In D. Pfoser (Ed.), Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings (pp. 422-440). Springer. Lecture Notes in Computer Science, Vol.. 6849 https://doi.org/10.1007/978-3-642-22922-0_25
Bernecker, Thomas ; Houle, Michael E. ; Kriegel, Hans Peter ; Kröger, Peer ; Renz, Matthias ; Schubert, Erich ; Zimek, Arthur. / Quality of similarity rankings in time series. Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings. editor / D. Pfoser. Springer, 2011. pp. 422-440 (Lecture Notes in Computer Science, Vol. 6849).
@inproceedings{86bf4b2f1a6f424186bf0aa95bce1e27,
title = "Quality of similarity rankings in time series",
abstract = "Time series data objects can be interpreted as high- dimensional vectors, which allows the application of many traditional distance measures as well as more specialized measures. However, many distance functions are known to suffer from poor contrast in high-dimensional settings, putting their usefulness as similarity measures into question. On the other hand, shared-nearest-neighbor distances based on the ranking of data objects induced by some primary distance measure have been known to lead to improved performance in high-dimensional settings. In this paper, we study the performance of shared-neighbor similarity measures in the context of similarity search for time series data objects. Our findings are that the use of shared-neighbor similarity measures generally results in more stable performances than that of their associated primary distance measures.",
author = "Thomas Bernecker and Houle, {Michael E.} and Kriegel, {Hans Peter} and Peer Kr{\"o}ger and Matthias Renz and Erich Schubert and Arthur Zimek",
year = "2011",
month = "9",
day = "19",
doi = "10.1007/978-3-642-22922-0_25",
language = "English",
isbn = "978-3-642-22921-3",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "422--440",
editor = "D. Pfoser",
booktitle = "Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings",
address = "Germany",

}

Bernecker, T, Houle, ME, Kriegel, HP, Kröger, P, Renz, M, Schubert, E & Zimek, A 2011, Quality of similarity rankings in time series. in D Pfoser (ed.), Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings. Springer, Lecture Notes in Computer Science, vol. 6849, pp. 422-440, 12th International Symposium on Advances in Spatial and Temporal Databases, Minneapolis, United States, 24/08/2011. https://doi.org/10.1007/978-3-642-22922-0_25

Quality of similarity rankings in time series. / Bernecker, Thomas; Houle, Michael E.; Kriegel, Hans Peter; Kröger, Peer; Renz, Matthias; Schubert, Erich; Zimek, Arthur.

Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings. ed. / D. Pfoser. Springer, 2011. p. 422-440 (Lecture Notes in Computer Science, Vol. 6849).

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

TY - GEN

T1 - Quality of similarity rankings in time series

AU - Bernecker, Thomas

AU - Houle, Michael E.

AU - Kriegel, Hans Peter

AU - Kröger, Peer

AU - Renz, Matthias

AU - Schubert, Erich

AU - Zimek, Arthur

PY - 2011/9/19

Y1 - 2011/9/19

N2 - Time series data objects can be interpreted as high- dimensional vectors, which allows the application of many traditional distance measures as well as more specialized measures. However, many distance functions are known to suffer from poor contrast in high-dimensional settings, putting their usefulness as similarity measures into question. On the other hand, shared-nearest-neighbor distances based on the ranking of data objects induced by some primary distance measure have been known to lead to improved performance in high-dimensional settings. In this paper, we study the performance of shared-neighbor similarity measures in the context of similarity search for time series data objects. Our findings are that the use of shared-neighbor similarity measures generally results in more stable performances than that of their associated primary distance measures.

AB - Time series data objects can be interpreted as high- dimensional vectors, which allows the application of many traditional distance measures as well as more specialized measures. However, many distance functions are known to suffer from poor contrast in high-dimensional settings, putting their usefulness as similarity measures into question. On the other hand, shared-nearest-neighbor distances based on the ranking of data objects induced by some primary distance measure have been known to lead to improved performance in high-dimensional settings. In this paper, we study the performance of shared-neighbor similarity measures in the context of similarity search for time series data objects. Our findings are that the use of shared-neighbor similarity measures generally results in more stable performances than that of their associated primary distance measures.

U2 - 10.1007/978-3-642-22922-0_25

DO - 10.1007/978-3-642-22922-0_25

M3 - Article in proceedings

AN - SCOPUS:80052770417

SN - 978-3-642-22921-3

T3 - Lecture Notes in Computer Science

SP - 422

EP - 440

BT - Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings

A2 - Pfoser, D.

PB - Springer

ER -

Bernecker T, Houle ME, Kriegel HP, Kröger P, Renz M, Schubert E et al. Quality of similarity rankings in time series. In Pfoser D, editor, Advances in Spatial and Temporal Databases - 12th International Symposium, SSTD 2011, Proceedings. Springer. 2011. p. 422-440. (Lecture Notes in Computer Science, Vol. 6849). https://doi.org/10.1007/978-3-642-22922-0_25