Three Myths about Dynamic Time Warping Data Mining

By Chotirat (Ann) Ratanamahatana, Eamonn J. Keogh. SDM, 2005

477 citations recorded by the source corpus.

The Dynamic Time Warping (DTW) distance measure is a technique that has long been known in speech recognition community. It allows a non-linear mapping of one signal to another by minimizing the distance between the two. A decade ago, DTW was introduced into Data Mining community as a utility for various tasks for time series problems including classification, clustering, and anomaly detection. The technique has flourished, particularly in the last three years, and has been applied to a variety of problems in various disciplines. In spite of DTW’s great success, there are still several persistent “myths” about it. These myths have caused confusion and led to much wasted research effort. In this work, we will dispel these myths with the most comprehensive set of time series experiments ever conducted.

Subfields: Biomedical Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/d140cb3e753eefc3d43cace8da5a5503aeb9aba1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Signal Processing

Citations in the corpus

Learning Context-Sensitive Shape Similarity by Graph Transduction (2010, methodology): “We used a dynamic time-warping (DTW) algorithm with a warping window [52] to generate the distance matrix, and obtained the first nearest neighbor (1NN) retrieval accuracy of 88.”
Data Augmentation for Time Series Classification using Convolutional Neural Networks (2016, methodology): “The combination of DTW and k -nearest-neighbors ( k -NN) is known to be a very e ﬃ cient e ﬃ cient approach for TS classiﬁcation [13].”
Characteristic-Based Clustering for Time Series Data (2006, background): “Dynamic Time Warping (DTW) has been applied in time series mining to resolve the difficulty in clustering time series of variable lengths in Euclidean space or containing possible out-of-phase similarities (Berndt and Clifford, 1994; Ratanamahatana and Keogh, 2005).”
A Bag-of-Features Framework to Classify Time Series (2013, background): “DTW [6] allows a measure of the similarity invariant to certain nonlinear variations in the time dimension and is considered as a strong solution for time series problems [7].”
Time series classification with ensembles of elastic distance measures (2015, methodology): “For similarity in shape, DTW is commonly used to mitigate against distortions in the time axis (Ratanamahatana and Keogh 2005).”
Deep learning for time series classification: a review (2018, methodology): “This form of pre-processing has also been used by Ratanamahatana and Keogh (2005) to show that the length of a time series is not an issue for TSC problems.”
Time-series clustering - A decade review (2015, background): “A research has revealed that similarity in shape is superior to metrics based on similarity in time [115].”
Exploiting multi-channels deep convolutional neural networks for multivariate time series classification (2015, unknown): “Hence, to reduce the time consumption, window constraint DTW has been adopted widely instead of full DTW in many previous work [10, 18–20].”
The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances (2020, methodology): “For the univariate archive, tuning the window parameter gives a small, but significant, overall improvement over using the full window size Ratanamahatana and Keogh (2005).”
Fast time series classification using numerosity reduction (2006, methodology): “To verify the claim, we experimented on the Gun-Point Dataset using one-nearest-neighbor DTW (a description of the dataset can be found in Ratanamahatana & Keogh, 2005).”
A time series forest for classification and feature extraction (2013, background): “Usually NNDTW performs better than NNEuclidean (dynamic time warping [17] is robust to the distortion in the time axis), and is considered as a strong solution for time series problems [13].”