QP-DTW: Upgrading Dynamic Time Warping to Handle Quasi Periodic Time Series Alignment
Dynamic time warping (DTW) is the main algorithms for time series alignment. However, it is unsuitable forquasi-periodic time series. In the current situation, except the recently published the shape exchangealgorithm (SEA) method and its derivatives, no other technique is able to handle alignment o...
Gespeichert in:
Veröffentlicht in: | Journal of information processing systems 2018, 14(4), 52, pp.851-876 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Dynamic time warping (DTW) is the main algorithms for time series alignment. However, it is unsuitable forquasi-periodic time series. In the current situation, except the recently published the shape exchangealgorithm (SEA) method and its derivatives, no other technique is able to handle alignment of this type ofvery complex time series. In this work, we propose a novel algorithm that combines the advantages of the SEAand the DTW methods. Our main contribution consists in the elevation of the DTW power of alignmentfrom the lowest level (Class A, non-periodic time series) to the highest level (Class C, multiple-periods timeseries containing different number of periods each), according to the recent classification of time seriesalignment methods proposed by Boucheham (Int J Mach Learn Cybern, vol. 4, no. 5, pp. 537-550, 2013). Thenew method (quasi-periodic dynamic time warping [QP-DTW]) was compared to both SEA and DTWmethods on electrocardiogram (ECG) time series, selected from the Massachusetts Institute of Technology -Beth Israel Hospital (MIT-BIH) public database and from the PTB Diagnostic ECG Database. Results showthat the proposed algorithm is more effective than DTW and SEA in terms of alignment accuracy on bothqualitative and quantitative levels. Therefore, QP-DTW would potentially be more suitable for manyapplications related to time series (e.g., data mining, pattern recognition, search/retrieval, motif discovery,classification, etc.). KCI Citation Count: 1 |
---|---|
ISSN: | 1976-913X 2092-805X |
DOI: | 10.3745/JIPS.02.0090 |