Twin Subsequence Search in Time Series
Data Structures and Algorithms
2021-04-15 v1
Abstract
We address the problem of subsequence search in time series using Chebyshev distance, to which we refer as twin subsequence search. We first show how existing time series indices can be extended to perform twin subsequence search. Then, we introduce TS-Index, a novel index tailored to this problem. Our experimental evaluation compares these approaches against real time series datasets, and demonstrates that TS-Index can retrieve twin subsequences much faster under various query conditions. This paper has been published in the 24th International Conference on Extending Database Technology (EDBT 2021).
Cite
@article{arxiv.2104.06874,
title = {Twin Subsequence Search in Time Series},
author = {Georgios Chatzigeorgakidis and Dimitrios Skoutas and Kostas Patroumpas and Themis Palpanas and Spiros Athanasiou and Spiros Skiadopoulos},
journal= {arXiv preprint arXiv:2104.06874},
year = {2021}
}
Comments
6 pages, 8 figures