DTW+S: Shape-based Comparison of Time-series with Ordered Local Trend

TL;DR

This paper introduces DTW+S, a novel, interpretable distance measure for time-series that emphasizes local trend similarities and ordered local patterns, improving clustering, classification, and ensemble methods.

Contribution

The paper proposes DTW+S, a new shape-based time-series similarity measure that captures ordered local trends and provides an interpretable matrix representation, supported by theoretical analysis.

Findings

DTW+S outperforms baselines in epidemic curve clustering.

Combining DTW+S with barycenter averaging enhances trajectory preservation.

DTW+S improves classification accuracy when local trends are crucial.

Abstract

Measuring distance or similarity between time-series data is a fundamental aspect of many applications including classification, clustering, and ensembling/alignment. Existing measures may fail to capture similarities among local trends (shapes) and may even produce misleading results. Our goal is to develop a measure that looks for similar trends occurring around similar times and is easily interpretable for researchers in applied domains. This is particularly useful for applications where time-series have a sequence of meaningful local trends that are ordered, such as in epidemics (a surge to an increase to a peak to a decrease). We propose a novel measure, DTW+S, which creates an interpretable "closeness-preserving" matrix representation of the time-series, where each column represents local trends, and then it applies Dynamic Time Warping to compute distances between these matrices. We present a theoretical analysis that supports the choice of this representation. We demonstrate the utility of DTW+S in several tasks. For the clustering of epidemic curves, we show that DTW+S is the only measure able to produce good clustering compared to the baselines. For ensemble building, we propose a combination of DTW+S and barycenter averaging that results in the best preservation of characteristics of the underlying trajectories. We also demonstrate that our approach results in better classification compared to Dynamic Time Warping for a class of datasets, particularly when local trends rather than scale play a decisive role.