Incremental methods for simple problems in time series: Algorithms and experiments

Xiaojian Zhao, Xin Zhang, Tyle Neylon, Dennis Shasha

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A time series (or equivalently a data stream) consists of data arriving in time order. Single or multiple data streams arise in fields including physics, finance, medicine, and music, to name a few. Often the data comes from sensors (in physics and medicine for example) whose data rates continue to improve dramatically as sensor technology improves and as the number of sensors increases. So fast algorithms become ever more critical in order to distill knowledge from the data. This paper presents our recent work regarding the incremental computation of various primitives: windowed correlation, matching pursuit, sparse null space discovery and elastic burst detection. The incremental idea reflects the fact that recent data is more important than older data. Our StatStream system contains an implementation of these algorithms, permitting us to do empirical studies on both simulated and real data.

Original languageEnglish (US)
Title of host publicationProceedings of the International Database Engineering and Applications Symposium, IDEAS
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3-14
Number of pages12
Volume2005-January
EditionJanuary
DOIs
StatePublished - 2005
Event9th International Database Engineering and Application Symposium, IDEAS 2005 - Montreal, Canada
Duration: Jul 25 2005Jul 27 2005

Other

Other9th International Database Engineering and Application Symposium, IDEAS 2005
CountryCanada
CityMontreal
Period7/25/057/27/05

Fingerprint

Time series
Medicine
Sensors
Physics
Experiments
Finance

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Zhao, X., Zhang, X., Neylon, T., & Shasha, D. (2005). Incremental methods for simple problems in time series: Algorithms and experiments. In Proceedings of the International Database Engineering and Applications Symposium, IDEAS (January ed., Vol. 2005-January, pp. 3-14). [1540890] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IDEAS.2005.35

Incremental methods for simple problems in time series : Algorithms and experiments. / Zhao, Xiaojian; Zhang, Xin; Neylon, Tyle; Shasha, Dennis.

Proceedings of the International Database Engineering and Applications Symposium, IDEAS. Vol. 2005-January January. ed. Institute of Electrical and Electronics Engineers Inc., 2005. p. 3-14 1540890.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhao, X, Zhang, X, Neylon, T & Shasha, D 2005, Incremental methods for simple problems in time series: Algorithms and experiments. in Proceedings of the International Database Engineering and Applications Symposium, IDEAS. January edn, vol. 2005-January, 1540890, Institute of Electrical and Electronics Engineers Inc., pp. 3-14, 9th International Database Engineering and Application Symposium, IDEAS 2005, Montreal, Canada, 7/25/05. https://doi.org/10.1109/IDEAS.2005.35
Zhao X, Zhang X, Neylon T, Shasha D. Incremental methods for simple problems in time series: Algorithms and experiments. In Proceedings of the International Database Engineering and Applications Symposium, IDEAS. January ed. Vol. 2005-January. Institute of Electrical and Electronics Engineers Inc. 2005. p. 3-14. 1540890 https://doi.org/10.1109/IDEAS.2005.35
Zhao, Xiaojian ; Zhang, Xin ; Neylon, Tyle ; Shasha, Dennis. / Incremental methods for simple problems in time series : Algorithms and experiments. Proceedings of the International Database Engineering and Applications Symposium, IDEAS. Vol. 2005-January January. ed. Institute of Electrical and Electronics Engineers Inc., 2005. pp. 3-14
@inproceedings{4fde57b4a74c4e409f50ec2664c58570,
title = "Incremental methods for simple problems in time series: Algorithms and experiments",
abstract = "A time series (or equivalently a data stream) consists of data arriving in time order. Single or multiple data streams arise in fields including physics, finance, medicine, and music, to name a few. Often the data comes from sensors (in physics and medicine for example) whose data rates continue to improve dramatically as sensor technology improves and as the number of sensors increases. So fast algorithms become ever more critical in order to distill knowledge from the data. This paper presents our recent work regarding the incremental computation of various primitives: windowed correlation, matching pursuit, sparse null space discovery and elastic burst detection. The incremental idea reflects the fact that recent data is more important than older data. Our StatStream system contains an implementation of these algorithms, permitting us to do empirical studies on both simulated and real data.",
author = "Xiaojian Zhao and Xin Zhang and Tyle Neylon and Dennis Shasha",
year = "2005",
doi = "10.1109/IDEAS.2005.35",
language = "English (US)",
volume = "2005-January",
pages = "3--14",
booktitle = "Proceedings of the International Database Engineering and Applications Symposium, IDEAS",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
edition = "January",

}

TY - GEN

T1 - Incremental methods for simple problems in time series

T2 - Algorithms and experiments

AU - Zhao, Xiaojian

AU - Zhang, Xin

AU - Neylon, Tyle

AU - Shasha, Dennis

PY - 2005

Y1 - 2005

N2 - A time series (or equivalently a data stream) consists of data arriving in time order. Single or multiple data streams arise in fields including physics, finance, medicine, and music, to name a few. Often the data comes from sensors (in physics and medicine for example) whose data rates continue to improve dramatically as sensor technology improves and as the number of sensors increases. So fast algorithms become ever more critical in order to distill knowledge from the data. This paper presents our recent work regarding the incremental computation of various primitives: windowed correlation, matching pursuit, sparse null space discovery and elastic burst detection. The incremental idea reflects the fact that recent data is more important than older data. Our StatStream system contains an implementation of these algorithms, permitting us to do empirical studies on both simulated and real data.

AB - A time series (or equivalently a data stream) consists of data arriving in time order. Single or multiple data streams arise in fields including physics, finance, medicine, and music, to name a few. Often the data comes from sensors (in physics and medicine for example) whose data rates continue to improve dramatically as sensor technology improves and as the number of sensors increases. So fast algorithms become ever more critical in order to distill knowledge from the data. This paper presents our recent work regarding the incremental computation of various primitives: windowed correlation, matching pursuit, sparse null space discovery and elastic burst detection. The incremental idea reflects the fact that recent data is more important than older data. Our StatStream system contains an implementation of these algorithms, permitting us to do empirical studies on both simulated and real data.

UR - http://www.scopus.com/inward/record.url?scp=78149353154&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149353154&partnerID=8YFLogxK

U2 - 10.1109/IDEAS.2005.35

DO - 10.1109/IDEAS.2005.35

M3 - Conference contribution

VL - 2005-January

SP - 3

EP - 14

BT - Proceedings of the International Database Engineering and Applications Symposium, IDEAS

PB - Institute of Electrical and Electronics Engineers Inc.

ER -