Abstract
Discovery of repeated patterns, known as motifs, from long time series is essential for providing hidden knowledge to real-world applications like medical, financial and weather analysis. Motifs can be discovered on raw time series directly or on their transformed abstract representation alternatively. Most of time series motif discovery methods require predefined motif length, which results in long execution time because we have to vary the length to discover motifs with different lengths. To solve the problem, we propose an efficient method for discovering variable length motifs in combination of approximate method with exact verification. First, symbolic representation is adopted to discover motifs roughly followed by exact examination of the found motifs with original real-valued data to achieve fast and exact discovery. The experiments show that our proposed method successfully discovered significant motifs efficiently in comparison with state-of-the-art methods: MK and SBF.
| Original language | English |
|---|---|
| Title of host publication | 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings |
| Publisher | Association for Computing Machinery |
| Pages | 49-57 |
| Number of pages | 9 |
| Volume | Part F134476 |
| ISBN (Electronic) | 9781450352994 |
| DOIs | |
| Publication status | Published - 2017 Dec 4 |
| Event | 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Salzburg, Austria Duration: 2017 Dec 4 → 2017 Dec 6 |
Other
| Other | 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 |
|---|---|
| Country | Austria |
| City | Salzburg |
| Period | 17/12/4 → 17/12/6 |
Fingerprint
Keywords
- Frequent pattern mining
- Motif
- Symbolic representation
- Time series
ASJC Scopus subject areas
- Human-Computer Interaction
- Computer Networks and Communications
- Computer Vision and Pattern Recognition
- Software
Cite this
A variable-length motifs discovery method in time series using hybrid approach. / Zan, Chaw Thet; Yamana, Hayato.
19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings. Vol. Part F134476 Association for Computing Machinery, 2017. p. 49-57.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - A variable-length motifs discovery method in time series using hybrid approach
AU - Zan, Chaw Thet
AU - Yamana, Hayato
PY - 2017/12/4
Y1 - 2017/12/4
N2 - Discovery of repeated patterns, known as motifs, from long time series is essential for providing hidden knowledge to real-world applications like medical, financial and weather analysis. Motifs can be discovered on raw time series directly or on their transformed abstract representation alternatively. Most of time series motif discovery methods require predefined motif length, which results in long execution time because we have to vary the length to discover motifs with different lengths. To solve the problem, we propose an efficient method for discovering variable length motifs in combination of approximate method with exact verification. First, symbolic representation is adopted to discover motifs roughly followed by exact examination of the found motifs with original real-valued data to achieve fast and exact discovery. The experiments show that our proposed method successfully discovered significant motifs efficiently in comparison with state-of-the-art methods: MK and SBF.
AB - Discovery of repeated patterns, known as motifs, from long time series is essential for providing hidden knowledge to real-world applications like medical, financial and weather analysis. Motifs can be discovered on raw time series directly or on their transformed abstract representation alternatively. Most of time series motif discovery methods require predefined motif length, which results in long execution time because we have to vary the length to discover motifs with different lengths. To solve the problem, we propose an efficient method for discovering variable length motifs in combination of approximate method with exact verification. First, symbolic representation is adopted to discover motifs roughly followed by exact examination of the found motifs with original real-valued data to achieve fast and exact discovery. The experiments show that our proposed method successfully discovered significant motifs efficiently in comparison with state-of-the-art methods: MK and SBF.
KW - Frequent pattern mining
KW - Motif
KW - Symbolic representation
KW - Time series
UR - http://www.scopus.com/inward/record.url?scp=85044283084&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044283084&partnerID=8YFLogxK
U2 - 10.1145/3151759.3151781
DO - 10.1145/3151759.3151781
M3 - Conference contribution
AN - SCOPUS:85044283084
VL - Part F134476
SP - 49
EP - 57
BT - 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings
PB - Association for Computing Machinery
ER -