Dimension reduction for clustering time series using global characteristics

Xiaozhe Wang; Kate A. Smith; Rob J. Hyndman

doi:10.1007/11428862_108

Dimension reduction for clustering time series using global characteristics

Xiaozhe Wang, Kate A. Smith, Rob J. Hyndman

Econometrics & Business Statistics

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

29 Citations (Scopus)

Abstract

Existing methods for time series clustering rely on the actual data values can become impractical since the methods do not easily handle dataset with high dimensionality, missing value, or different lengths. In this paper, a dimension reduction method is proposed that replaces the raw data with some global measures of time series characteristics. These measures are then clustered using a self-organizing map. The proposed approach has been tested using benchmark time series previously reported for time series clustering, and is shown to yield useful and robust clustering.

Original language	English
Title of host publication	Computational Science – ICCS 2005
Subtitle of host publication	5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I
Editors	Vaidy S. Sunderam, Geert Dick van Albada, Peter M.A. Sloot, Jack J. Dongarra
Place of Publication	Berlin Germany
Publisher	Springer
Pages	792-795
Number of pages	4
ISBN (Print)	3540260323, 9783540260325
DOIs	https://doi.org/10.1007/11428862_108
Publication status	Published - 30 Sept 2005
Event	International Conference on Computational Science 2005 - Atlanta, United States of America Duration: 22 May 2005 → 25 May 2005 Conference number: 5th https://link-springer-com.ezproxy.lib.monash.edu.au/book/10.1007/b136570 (Proceedings)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	3516
ISSN (Print)	0302-9743

Conference

Conference	International Conference on Computational Science 2005
Abbreviated title	ICCS 2005
Country/Territory	United States of America
City	Atlanta
Period	22/05/05 → 25/05/05
Internet address	https://link-springer-com.ezproxy.lib.monash.edu.au/book/10.1007/b136570 (Proceedings)

Access to Document

10.1007/11428862_108

Cite this

Wang, X., Smith, K. A., & Hyndman, R. J. (2005). Dimension reduction for clustering time series using global characteristics. In V. S. Sunderam, G. D. van Albada, P. M. A. Sloot, & J. J. Dongarra (Eds.), Computational Science – ICCS 2005: 5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I (pp. 792-795). (Lecture Notes in Computer Science; Vol. 3516). Springer. https://doi.org/10.1007/11428862_108

Wang, Xiaozhe ; Smith, Kate A. ; Hyndman, Rob J. / Dimension reduction for clustering time series using global characteristics. Computational Science – ICCS 2005: 5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I. editor / Vaidy S. Sunderam ; Geert Dick van Albada ; Peter M.A. Sloot ; Jack J. Dongarra. Berlin Germany : Springer, 2005. pp. 792-795 (Lecture Notes in Computer Science).

@inproceedings{52621b80f0114fd492d6e82735ff8867,

title = "Dimension reduction for clustering time series using global characteristics",

abstract = "Existing methods for time series clustering rely on the actual data values can become impractical since the methods do not easily handle dataset with high dimensionality, missing value, or different lengths. In this paper, a dimension reduction method is proposed that replaces the raw data with some global measures of time series characteristics. These measures are then clustered using a self-organizing map. The proposed approach has been tested using benchmark time series previously reported for time series clustering, and is shown to yield useful and robust clustering.",

author = "Xiaozhe Wang and Smith, {Kate A.} and Hyndman, {Rob J.}",

year = "2005",

month = sep,

day = "30",

doi = "10.1007/11428862_108",

language = "English",

isbn = "3540260323",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "792--795",

editor = "Sunderam, {Vaidy S.} and {van Albada}, {Geert Dick} and Sloot, {Peter M.A.} and Dongarra, { Jack J.}",

booktitle = "Computational Science – ICCS 2005",

note = "International Conference on Computational Science 2005, ICCS 2005 ; Conference date: 22-05-2005 Through 25-05-2005",

url = "https://link-springer-com.ezproxy.lib.monash.edu.au/book/10.1007/b136570",

}

Wang, X, Smith, KA & Hyndman, RJ 2005, Dimension reduction for clustering time series using global characteristics. in VS Sunderam, GD van Albada, PMA Sloot & JJ Dongarra (eds), Computational Science – ICCS 2005: 5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I. Lecture Notes in Computer Science, vol. 3516, Springer, Berlin Germany, pp. 792-795, International Conference on Computational Science 2005, Atlanta, Georgia, United States of America, 22/05/05. https://doi.org/10.1007/11428862_108

Dimension reduction for clustering time series using global characteristics. / Wang, Xiaozhe; Smith, Kate A.; Hyndman, Rob J.
Computational Science – ICCS 2005: 5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I. ed. / Vaidy S. Sunderam; Geert Dick van Albada; Peter M.A. Sloot; Jack J. Dongarra. Berlin Germany: Springer, 2005. p. 792-795 (Lecture Notes in Computer Science; Vol. 3516).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Dimension reduction for clustering time series using global characteristics

AU - Wang, Xiaozhe

AU - Smith, Kate A.

AU - Hyndman, Rob J.

N1 - Conference code: 5th

PY - 2005/9/30

Y1 - 2005/9/30

N2 - Existing methods for time series clustering rely on the actual data values can become impractical since the methods do not easily handle dataset with high dimensionality, missing value, or different lengths. In this paper, a dimension reduction method is proposed that replaces the raw data with some global measures of time series characteristics. These measures are then clustered using a self-organizing map. The proposed approach has been tested using benchmark time series previously reported for time series clustering, and is shown to yield useful and robust clustering.

AB - Existing methods for time series clustering rely on the actual data values can become impractical since the methods do not easily handle dataset with high dimensionality, missing value, or different lengths. In this paper, a dimension reduction method is proposed that replaces the raw data with some global measures of time series characteristics. These measures are then clustered using a self-organizing map. The proposed approach has been tested using benchmark time series previously reported for time series clustering, and is shown to yield useful and robust clustering.

UR - http://www.scopus.com/inward/record.url?scp=25144474901&partnerID=8YFLogxK

U2 - 10.1007/11428862_108

DO - 10.1007/11428862_108

M3 - Conference Paper

AN - SCOPUS:25144474901

SN - 3540260323

SN - 9783540260325

T3 - Lecture Notes in Computer Science

SP - 792

EP - 795

BT - Computational Science – ICCS 2005

A2 - Sunderam, Vaidy S.

A2 - van Albada, Geert Dick

A2 - Sloot, Peter M.A.

A2 - Dongarra, Jack J.

PB - Springer

CY - Berlin Germany

T2 - International Conference on Computational Science 2005

Y2 - 22 May 2005 through 25 May 2005

ER -

Wang X, Smith KA, Hyndman RJ. Dimension reduction for clustering time series using global characteristics. In Sunderam VS, van Albada GD, Sloot PMA, Dongarra JJ, editors, Computational Science – ICCS 2005: 5th International Conference Atlanta, GA, USA, May 22-25, 2005 Proceedings, Part I. Berlin Germany: Springer. 2005. p. 792-795. (Lecture Notes in Computer Science). doi: 10.1007/11428862_108

Dimension reduction for clustering time series using global characteristics

Abstract

Publication series

Conference

Access to Document

Other files and links

Cite this