DMCA
FTW: Fast Similarity Search under the Time Warping (2005)
| Venue: | Distance, Proceedings of PODS |
| Citations: | 29 - 2 self |
BibTeX
@INPROCEEDINGS{Sakurai05ftw:fast,
author = {Yasushi Sakurai and Masatoshi Yoshikawa and Christos Faloutsos},
title = {FTW: Fast Similarity Search under the Time Warping},
booktitle = {Distance, Proceedings of PODS},
year = {2005}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Time-series data naturally arise in countless domains, such as meteorology, astrophysics, geology, multimedia, and eco-nomics. Similarity search is very popular, and DTW (Dy-namic Time Warping) is one of the two prevailing distance measures. Although DTW incurs a heavy computation cost, it provides scaling along the time axis. In this paper, we pro-pose FTW (Fast search method for dynamic Time Warping), which guarantees no false dismissals in similarity query pro-cessing. FTW efficiently prunes a significant number of the search candidates, which leads to a direct reduction in the search cost. Experiments on real and synthetic sequence data sets reveal that FTW is significantly faster than the best existing method, up to 222 times. 1.
Keyphrases
time warping fast similarity search dynamic time warping time-series data significant number time axis similarity query pro-cessing distance measure search cost heavy computation cost search candidate synthetic sequence data set similarity search direct reduction false dismissal dy-namic time warping fast search method countless domain
