Abstract

Many applications generate and/or consume multi-variate temporal data, yet experts often lack the means to adequately and systematically search for and interpret multi-variate observations. In this paper, we first observe that multi-variate time series often carry localized multi-variate temporal features that are robust against noise. We then argue that these multi-variate temporal features can be extracted by simultaneously considering, at multiple scales, temporal characteristics of the time-series along with external knowledge, including variate relationships, known a priori. Relying on these observations, we develop algorithms to detect robust multi-variate temporal (RMT) features which can be indexed for efficient and accurate retrieval and can be used for supporting analysis tasks, such as classification. Experiments confirm that the proposed RMT algorithm is highly effective and efficient in identifying robust multi-scale temporal features of multi-variate time series.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Data Engineering
PublisherIEEE Computer Society
Pages388-399
Number of pages12
ISBN (Print)9781479925544
DOIs
StatePublished - 2014
Event30th IEEE International Conference on Data Engineering, ICDE 2014 - Chicago, IL, United States
Duration: Mar 31 2014Apr 4 2014

Other

Other30th IEEE International Conference on Data Engineering, ICDE 2014
CountryUnited States
CityChicago, IL
Period3/31/144/4/14

Fingerprint

Metadata
Time series
Experiments

ASJC Scopus subject areas

  • Information Systems
  • Signal Processing
  • Software

Cite this

Wang, X., Candan, K., & Sapino, M. L. (2014). Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features. In Proceedings - International Conference on Data Engineering (pp. 388-399). [6816667] IEEE Computer Society. https://doi.org/10.1109/ICDE.2014.6816667

Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features. / Wang, Xiaolan; Candan, Kasim; Sapino, Maria Luisa.

Proceedings - International Conference on Data Engineering. IEEE Computer Society, 2014. p. 388-399 6816667.

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

Wang, X, Candan, K & Sapino, ML 2014, Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features. in Proceedings - International Conference on Data Engineering., 6816667, IEEE Computer Society, pp. 388-399, 30th IEEE International Conference on Data Engineering, ICDE 2014, Chicago, IL, United States, 3/31/14. https://doi.org/10.1109/ICDE.2014.6816667
Wang X, Candan K, Sapino ML. Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features. In Proceedings - International Conference on Data Engineering. IEEE Computer Society. 2014. p. 388-399. 6816667 https://doi.org/10.1109/ICDE.2014.6816667
Wang, Xiaolan ; Candan, Kasim ; Sapino, Maria Luisa. / Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features. Proceedings - International Conference on Data Engineering. IEEE Computer Society, 2014. pp. 388-399
@inproceedings{3efcfea407d64b808a93dbc254264322,
title = "Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features",
abstract = "Many applications generate and/or consume multi-variate temporal data, yet experts often lack the means to adequately and systematically search for and interpret multi-variate observations. In this paper, we first observe that multi-variate time series often carry localized multi-variate temporal features that are robust against noise. We then argue that these multi-variate temporal features can be extracted by simultaneously considering, at multiple scales, temporal characteristics of the time-series along with external knowledge, including variate relationships, known a priori. Relying on these observations, we develop algorithms to detect robust multi-variate temporal (RMT) features which can be indexed for efficient and accurate retrieval and can be used for supporting analysis tasks, such as classification. Experiments confirm that the proposed RMT algorithm is highly effective and efficient in identifying robust multi-scale temporal features of multi-variate time series.",
author = "Xiaolan Wang and Kasim Candan and Sapino, {Maria Luisa}",
year = "2014",
doi = "10.1109/ICDE.2014.6816667",
language = "English (US)",
isbn = "9781479925544",
pages = "388--399",
booktitle = "Proceedings - International Conference on Data Engineering",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Leveraging metadata for identifying local, robust multi-variate temporal (RMT) features

AU - Wang, Xiaolan

AU - Candan, Kasim

AU - Sapino, Maria Luisa

PY - 2014

Y1 - 2014

N2 - Many applications generate and/or consume multi-variate temporal data, yet experts often lack the means to adequately and systematically search for and interpret multi-variate observations. In this paper, we first observe that multi-variate time series often carry localized multi-variate temporal features that are robust against noise. We then argue that these multi-variate temporal features can be extracted by simultaneously considering, at multiple scales, temporal characteristics of the time-series along with external knowledge, including variate relationships, known a priori. Relying on these observations, we develop algorithms to detect robust multi-variate temporal (RMT) features which can be indexed for efficient and accurate retrieval and can be used for supporting analysis tasks, such as classification. Experiments confirm that the proposed RMT algorithm is highly effective and efficient in identifying robust multi-scale temporal features of multi-variate time series.

AB - Many applications generate and/or consume multi-variate temporal data, yet experts often lack the means to adequately and systematically search for and interpret multi-variate observations. In this paper, we first observe that multi-variate time series often carry localized multi-variate temporal features that are robust against noise. We then argue that these multi-variate temporal features can be extracted by simultaneously considering, at multiple scales, temporal characteristics of the time-series along with external knowledge, including variate relationships, known a priori. Relying on these observations, we develop algorithms to detect robust multi-variate temporal (RMT) features which can be indexed for efficient and accurate retrieval and can be used for supporting analysis tasks, such as classification. Experiments confirm that the proposed RMT algorithm is highly effective and efficient in identifying robust multi-scale temporal features of multi-variate time series.

UR - http://www.scopus.com/inward/record.url?scp=84901764871&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901764871&partnerID=8YFLogxK

U2 - 10.1109/ICDE.2014.6816667

DO - 10.1109/ICDE.2014.6816667

M3 - Conference contribution

AN - SCOPUS:84901764871

SN - 9781479925544

SP - 388

EP - 399

BT - Proceedings - International Conference on Data Engineering

PB - IEEE Computer Society

ER -