Projects per year
Abstract
Shapelets have recently been proposed as a new primitive for time series classification. Shapelets are subseries of series that best split the data into its classes. In the original research, shapelets were found recursively within a decision tree through enumeration of the search space. Subsequent research indicated that using shapelets as the basis for transforming datasets leads to more accurate classifiers.
Both these approaches evaluate how well a shapelet splits all the classes. However, often a shapelet is most useful in distinguishing between members of the class of the series it was drawn from against all others. To assess this conjecture, we evaluate a one vs all encoding scheme. This technique simplifies the quality assessment calculations, speeds up the execution through facilitating more frequent early abandon and increases accuracy for multi-class problems. We also propose an alternative shapelet evaluation scheme which we demonstrate significantly speeds up the full search.
Both these approaches evaluate how well a shapelet splits all the classes. However, often a shapelet is most useful in distinguishing between members of the class of the series it was drawn from against all others. To assess this conjecture, we evaluate a one vs all encoding scheme. This technique simplifies the quality assessment calculations, speeds up the execution through facilitating more frequent early abandon and increases accuracy for multi-class problems. We also propose an alternative shapelet evaluation scheme which we demonstrate significantly speeds up the full search.
Original language | English |
---|---|
Title of host publication | Big Data Analytics and Knowledge Discovery |
Subtitle of host publication | 17th International Conference, DaWaK 2015, Valencia, Spain, September 1-4, 2015, Proceedings |
Editors | Sanjay Madria, Takahiro Hara |
Publisher | Springer |
Pages | 257-269 |
Volume | 9263 |
ISBN (Electronic) | 978-3-319-22729-0 |
ISBN (Print) | 978-3-319-22728-3 |
DOIs | |
Publication status | Published - 1 Sept 2015 |
Event | DaWaK 2015 : 17th International Conference on Big Data Analytics and Knowledge Discovery - Valencia, Spain Duration: 1 Sept 2015 → 4 Sept 2015 |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Publisher | Springer |
Volume | 9263 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | DaWaK 2015 : 17th International Conference on Big Data Analytics and Knowledge Discovery |
---|---|
Country/Territory | Spain |
City | Valencia |
Period | 1/09/15 → 4/09/15 |
Keywords
- shapelets
- Time Series Classification
Profiles
Projects
- 1 Finished
-
The Collective of Transform Ensembles (COTE) for Time Series Classification
Engineering and Physical Sciences Research Council
1/05/15 → 31/10/18
Project: Research
Research output
- 51 Citations (Scopus)
- 1 Article
-
The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances
Bagnall, A., Lines, J., Bostrom, A., Large, J. & Keogh, E., May 2017, In: Data Mining and Knowledge Discovery. 31, 3, p. 606–660 55 p.Research output: Contribution to journal › Article › peer-review
Open AccessFile1031 Citations (SciVal)102 Downloads (Pure)