Multivariate time series classification with temporal abstractions

Lyad Batal, Lucia Sacchi, Riccardo Bellazzi, Milos Hauskrecht

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

Abstract

The increase in the number of complex temporal datasets collected today has prompted the development of methods that extend classical machine learning and data mining methods to time-series data. This work focuses on methods for multivariate time-series classification. Time series classification is a challenging problem mostly because the number of temporal features that describe the data and are potentially useful for classification is enormous. We study and develop a temporal abstraction framework for generating multivariate time series features suitable for classification tasks. We propose the STF-Mine algorithm that automatically mines discriminative temporal abstraction patterns from the time series data and uses them to learn a classification model. Our experimental evaluations, carried out on both synthetic and real world medical data, demonstrate the benefit of our approach in learning accurate classifiers for time-series datasets.

Original languageEnglish
Title of host publicationProceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22
Pages344-349
Number of pages6
Publication statusPublished - 2009
Event22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22 - Sanibel Island, FL, United States
Duration: Mar 19 2009Mar 21 2009

Other

Other22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22
CountryUnited States
CitySanibel Island, FL
Period3/19/093/21/09

Fingerprint

Time series
Data mining
Learning systems
Classifiers

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Software

Cite this

Batal, L., Sacchi, L., Bellazzi, R., & Hauskrecht, M. (2009). Multivariate time series classification with temporal abstractions. In Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22 (pp. 344-349)

Multivariate time series classification with temporal abstractions. / Batal, Lyad; Sacchi, Lucia; Bellazzi, Riccardo; Hauskrecht, Milos.

Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22. 2009. p. 344-349.

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

Batal, L, Sacchi, L, Bellazzi, R & Hauskrecht, M 2009, Multivariate time series classification with temporal abstractions. in Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22. pp. 344-349, 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22, Sanibel Island, FL, United States, 3/19/09.
Batal L, Sacchi L, Bellazzi R, Hauskrecht M. Multivariate time series classification with temporal abstractions. In Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22. 2009. p. 344-349
Batal, Lyad ; Sacchi, Lucia ; Bellazzi, Riccardo ; Hauskrecht, Milos. / Multivariate time series classification with temporal abstractions. Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22. 2009. pp. 344-349
@inproceedings{3db10fef88ee42b9b9b9257b4a6d0878,
title = "Multivariate time series classification with temporal abstractions",
abstract = "The increase in the number of complex temporal datasets collected today has prompted the development of methods that extend classical machine learning and data mining methods to time-series data. This work focuses on methods for multivariate time-series classification. Time series classification is a challenging problem mostly because the number of temporal features that describe the data and are potentially useful for classification is enormous. We study and develop a temporal abstraction framework for generating multivariate time series features suitable for classification tasks. We propose the STF-Mine algorithm that automatically mines discriminative temporal abstraction patterns from the time series data and uses them to learn a classification model. Our experimental evaluations, carried out on both synthetic and real world medical data, demonstrate the benefit of our approach in learning accurate classifiers for time-series datasets.",
author = "Lyad Batal and Lucia Sacchi and Riccardo Bellazzi and Milos Hauskrecht",
year = "2009",
language = "English",
isbn = "9781577354192",
pages = "344--349",
booktitle = "Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22",

}

TY - GEN

T1 - Multivariate time series classification with temporal abstractions

AU - Batal, Lyad

AU - Sacchi, Lucia

AU - Bellazzi, Riccardo

AU - Hauskrecht, Milos

PY - 2009

Y1 - 2009

N2 - The increase in the number of complex temporal datasets collected today has prompted the development of methods that extend classical machine learning and data mining methods to time-series data. This work focuses on methods for multivariate time-series classification. Time series classification is a challenging problem mostly because the number of temporal features that describe the data and are potentially useful for classification is enormous. We study and develop a temporal abstraction framework for generating multivariate time series features suitable for classification tasks. We propose the STF-Mine algorithm that automatically mines discriminative temporal abstraction patterns from the time series data and uses them to learn a classification model. Our experimental evaluations, carried out on both synthetic and real world medical data, demonstrate the benefit of our approach in learning accurate classifiers for time-series datasets.

AB - The increase in the number of complex temporal datasets collected today has prompted the development of methods that extend classical machine learning and data mining methods to time-series data. This work focuses on methods for multivariate time-series classification. Time series classification is a challenging problem mostly because the number of temporal features that describe the data and are potentially useful for classification is enormous. We study and develop a temporal abstraction framework for generating multivariate time series features suitable for classification tasks. We propose the STF-Mine algorithm that automatically mines discriminative temporal abstraction patterns from the time series data and uses them to learn a classification model. Our experimental evaluations, carried out on both synthetic and real world medical data, demonstrate the benefit of our approach in learning accurate classifiers for time-series datasets.

UR - http://www.scopus.com/inward/record.url?scp=70350496548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350496548&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:70350496548

SN - 9781577354192

SP - 344

EP - 349

BT - Proceedings of the 22nd International Florida Artificial Intelligence Research Society Conference, FLAIRS-22

ER -