Change-point detection in time-series data based on subspace identification

Yoshinobu Kawahara, Takehisa Yairi, Kazuo Machida

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

43 Citations (Scopus)

Abstract

In this paper, we propose series of algorithms for detecting change points in time-series data based on subspace identification, meaning a geometric approach for estimating linear state-space models behind time-series data. Our algorithms are derived from the principle that the subspace spanned by the columns of an observability matrix and the one spanned by the subsequences of time-series data are approximately equivalent. In this paper, we derive an batch-type algorithm applicable to ordinary time-series data, i.e. consisting of only output series, and then introduce the online version of the algorithm and the extension to be available with input-output time-series data. We illustrate the effectiveness of our algorithms with comparative experiments using some artificial and real datasets.

Original languageEnglish
Title of host publicationProceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007
Pages559-564
Number of pages6
DOIs
Publication statusPublished - Dec 1 2007
Externally publishedYes
Event7th IEEE International Conference on Data Mining, ICDM 2007 - Omaha, NE, United States
Duration: Oct 28 2007Oct 31 2007

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
ISSN (Print)1550-4786

Conference

Conference7th IEEE International Conference on Data Mining, ICDM 2007
CountryUnited States
CityOmaha, NE
Period10/28/0710/31/07

Fingerprint

Time series
Observability
Experiments

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Kawahara, Y., Yairi, T., & Machida, K. (2007). Change-point detection in time-series data based on subspace identification. In Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007 (pp. 559-564). [4470290] (Proceedings - IEEE International Conference on Data Mining, ICDM). https://doi.org/10.1109/ICDM.2007.78

Change-point detection in time-series data based on subspace identification. / Kawahara, Yoshinobu; Yairi, Takehisa; Machida, Kazuo.

Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. p. 559-564 4470290 (Proceedings - IEEE International Conference on Data Mining, ICDM).

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

Kawahara, Y, Yairi, T & Machida, K 2007, Change-point detection in time-series data based on subspace identification. in Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007., 4470290, Proceedings - IEEE International Conference on Data Mining, ICDM, pp. 559-564, 7th IEEE International Conference on Data Mining, ICDM 2007, Omaha, NE, United States, 10/28/07. https://doi.org/10.1109/ICDM.2007.78
Kawahara Y, Yairi T, Machida K. Change-point detection in time-series data based on subspace identification. In Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. p. 559-564. 4470290. (Proceedings - IEEE International Conference on Data Mining, ICDM). https://doi.org/10.1109/ICDM.2007.78
Kawahara, Yoshinobu ; Yairi, Takehisa ; Machida, Kazuo. / Change-point detection in time-series data based on subspace identification. Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. pp. 559-564 (Proceedings - IEEE International Conference on Data Mining, ICDM).
@inproceedings{c7e072ff27c2427bbe745c8e6206e269,
title = "Change-point detection in time-series data based on subspace identification",
abstract = "In this paper, we propose series of algorithms for detecting change points in time-series data based on subspace identification, meaning a geometric approach for estimating linear state-space models behind time-series data. Our algorithms are derived from the principle that the subspace spanned by the columns of an observability matrix and the one spanned by the subsequences of time-series data are approximately equivalent. In this paper, we derive an batch-type algorithm applicable to ordinary time-series data, i.e. consisting of only output series, and then introduce the online version of the algorithm and the extension to be available with input-output time-series data. We illustrate the effectiveness of our algorithms with comparative experiments using some artificial and real datasets.",
author = "Yoshinobu Kawahara and Takehisa Yairi and Kazuo Machida",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/ICDM.2007.78",
language = "English",
isbn = "0769530184",
series = "Proceedings - IEEE International Conference on Data Mining, ICDM",
pages = "559--564",
booktitle = "Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007",

}

TY - GEN

T1 - Change-point detection in time-series data based on subspace identification

AU - Kawahara, Yoshinobu

AU - Yairi, Takehisa

AU - Machida, Kazuo

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In this paper, we propose series of algorithms for detecting change points in time-series data based on subspace identification, meaning a geometric approach for estimating linear state-space models behind time-series data. Our algorithms are derived from the principle that the subspace spanned by the columns of an observability matrix and the one spanned by the subsequences of time-series data are approximately equivalent. In this paper, we derive an batch-type algorithm applicable to ordinary time-series data, i.e. consisting of only output series, and then introduce the online version of the algorithm and the extension to be available with input-output time-series data. We illustrate the effectiveness of our algorithms with comparative experiments using some artificial and real datasets.

AB - In this paper, we propose series of algorithms for detecting change points in time-series data based on subspace identification, meaning a geometric approach for estimating linear state-space models behind time-series data. Our algorithms are derived from the principle that the subspace spanned by the columns of an observability matrix and the one spanned by the subsequences of time-series data are approximately equivalent. In this paper, we derive an batch-type algorithm applicable to ordinary time-series data, i.e. consisting of only output series, and then introduce the online version of the algorithm and the extension to be available with input-output time-series data. We illustrate the effectiveness of our algorithms with comparative experiments using some artificial and real datasets.

UR - http://www.scopus.com/inward/record.url?scp=49749093757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=49749093757&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2007.78

DO - 10.1109/ICDM.2007.78

M3 - Conference contribution

AN - SCOPUS:49749093757

SN - 0769530184

SN - 9780769530185

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 559

EP - 564

BT - Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007

ER -