Improving the Efficiency of Traditional DTW Accelerators

Romain Tavenard 1 Laurent Amsaleg 2
2 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Dynamic Time Warping (DTW) is the most popular approach for evaluating the similarity of time series, but its computation is costly. Therefore, simple functions lower bounding DTW distances have been designed, accelerating searches by quickly pruning sequences that could not possibly be best matches. The tighter the bounds, the more they prune and the better the performance. Designing new functions that are even tighter is di fficult because their computation is likely to become complex, canceling the benefits of their pruning. It is possible, however, to design simple functions with a higher pruning power by relaxing the no false dismissal assumption, resulting in approximate lower bound functions. This paper describes how very popular approaches accelerating DTW such as LB Keogh and LB PAA can be made more e fficient via approximations. The accuracy of approximations can be tuned, ranging from no false dismissal to potential losses when aggressively set for great response time savings. At very large scale, indexing time series is mandatory. This paper also describes how approximate lower bound functions can be used with iSAX. Furthermore, it shows that a k-means-based quantization step for iSAX gives signi ficant performance gains.
Document type :
Journal articles
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal.inria.fr/hal-00862176
Contributor : Laurent Amsaleg <>
Submitted on : Wednesday, February 22, 2017 - 4:01:09 PM
Last modification on : Thursday, February 7, 2019 - 5:55:48 PM
Long-term archiving on : Tuesday, May 23, 2017 - 2:13:43 PM

File

hal-00862176.pdf
Files produced by the author(s)

Identifiers

Citation

Romain Tavenard, Laurent Amsaleg. Improving the Efficiency of Traditional DTW Accelerators. Knowledge and Information Systems (KAIS), Springer, 2015, 42 (1), pp.215-243. ⟨http://link.springer.com/article/10.1007/s10115-013-0698-7⟩. ⟨10.1007/s10115-013-0698-7⟩. ⟨hal-00862176⟩

Share

Metrics

Record views

440

Files downloads

197