Experimental evaluation of time-series decision tree

Yuu Yamada; Einoshin Suzuki; Hideto Yokoi; Katsuhiko Takabayashi

doi:10.1007/11423270_11

Experimental evaluation of time-series decision tree

Yuu Yamada, Einoshin Suzuki, Hideto Yokoi, Katsuhiko Takabayashi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

In this paper, we give experimental evaluation of our time-series decision tree induction method under various conditions. Our time-series tree has a value (i.e. a time sequence) of a time-series attribute in its internal node, and splits examples based on dissimilarity between a pair of time sequences. Our method selects, for a split test, a time sequence which exists in data by exhaustive search based on class and shape information. It has been empirically observed that the method induces accurate and comprehensive decision trees in time-series classification, which has gaining increasing attention due to its importance in various real-world applications. The evaluation has revealed several important findings including interaction between a split test and its measure of goodness.

Original language	English
Title of host publication	Active Mining - Second International Workshop, AM 2003, Revised Selected Papers
Publisher	Springer Verlag
Pages	190-209
Number of pages	20
ISBN (Print)	3540261575, 9783540261575
DOIs	https://doi.org/10.1007/11423270_11
Publication status	Published - 2005
Externally published	Yes
Event	Second International Workshop on Active Mining, AM 2003 - Maebashi, Japan Duration: Oct 28 2003 → Oct 31 2003

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	3430 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	Second International Workshop on Active Mining, AM 2003
Country/Territory	Japan
City	Maebashi
Period	10/28/03 → 10/31/03

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/11423270_11

Cite this

Yamada, Y., Suzuki, E., Yokoi, H., & Takabayashi, K. (2005). Experimental evaluation of time-series decision tree. In Active Mining - Second International Workshop, AM 2003, Revised Selected Papers (pp. 190-209). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3430 LNAI). Springer Verlag. https://doi.org/10.1007/11423270_11

Experimental evaluation of time-series decision tree. / Yamada, Yuu; Suzuki, Einoshin; Yokoi, Hideto et al.
Active Mining - Second International Workshop, AM 2003, Revised Selected Papers. Springer Verlag, 2005. p. 190-209 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3430 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yamada, Y, Suzuki, E, Yokoi, H & Takabayashi, K 2005, Experimental evaluation of time-series decision tree. in Active Mining - Second International Workshop, AM 2003, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3430 LNAI, Springer Verlag, pp. 190-209, Second International Workshop on Active Mining, AM 2003, Maebashi, Japan, 10/28/03. https://doi.org/10.1007/11423270_11

Yamada Y, Suzuki E, Yokoi H, Takabayashi K. Experimental evaluation of time-series decision tree. In Active Mining - Second International Workshop, AM 2003, Revised Selected Papers. Springer Verlag. 2005. p. 190-209. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/11423270_11

@inproceedings{d9a22958deb34d1eac81816cfa5014ca,

title = "Experimental evaluation of time-series decision tree",

abstract = "In this paper, we give experimental evaluation of our time-series decision tree induction method under various conditions. Our time-series tree has a value (i.e. a time sequence) of a time-series attribute in its internal node, and splits examples based on dissimilarity between a pair of time sequences. Our method selects, for a split test, a time sequence which exists in data by exhaustive search based on class and shape information. It has been empirically observed that the method induces accurate and comprehensive decision trees in time-series classification, which has gaining increasing attention due to its importance in various real-world applications. The evaluation has revealed several important findings including interaction between a split test and its measure of goodness.",

author = "Yuu Yamada and Einoshin Suzuki and Hideto Yokoi and Katsuhiko Takabayashi",

year = "2005",

doi = "10.1007/11423270_11",

language = "English",

isbn = "3540261575",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "190--209",

booktitle = "Active Mining - Second International Workshop, AM 2003, Revised Selected Papers",

address = "Germany",

note = "Second International Workshop on Active Mining, AM 2003 ; Conference date: 28-10-2003 Through 31-10-2003",

}

TY - GEN

T1 - Experimental evaluation of time-series decision tree

AU - Yamada, Yuu

AU - Suzuki, Einoshin

AU - Yokoi, Hideto

AU - Takabayashi, Katsuhiko

PY - 2005

Y1 - 2005

N2 - In this paper, we give experimental evaluation of our time-series decision tree induction method under various conditions. Our time-series tree has a value (i.e. a time sequence) of a time-series attribute in its internal node, and splits examples based on dissimilarity between a pair of time sequences. Our method selects, for a split test, a time sequence which exists in data by exhaustive search based on class and shape information. It has been empirically observed that the method induces accurate and comprehensive decision trees in time-series classification, which has gaining increasing attention due to its importance in various real-world applications. The evaluation has revealed several important findings including interaction between a split test and its measure of goodness.

AB - In this paper, we give experimental evaluation of our time-series decision tree induction method under various conditions. Our time-series tree has a value (i.e. a time sequence) of a time-series attribute in its internal node, and splits examples based on dissimilarity between a pair of time sequences. Our method selects, for a split test, a time sequence which exists in data by exhaustive search based on class and shape information. It has been empirically observed that the method induces accurate and comprehensive decision trees in time-series classification, which has gaining increasing attention due to its importance in various real-world applications. The evaluation has revealed several important findings including interaction between a split test and its measure of goodness.

UR - http://www.scopus.com/inward/record.url?scp=26844517756&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26844517756&partnerID=8YFLogxK

U2 - 10.1007/11423270_11

DO - 10.1007/11423270_11

M3 - Conference contribution

AN - SCOPUS:26844517756

SN - 3540261575

SN - 9783540261575

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 190

EP - 209

BT - Active Mining - Second International Workshop, AM 2003, Revised Selected Papers

PB - Springer Verlag

T2 - Second International Workshop on Active Mining, AM 2003

Y2 - 28 October 2003 through 31 October 2003

ER -

Experimental evaluation of time-series decision tree

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this