A similarity measure of Jumping Dynamic Time Warping

The problem of similarity measure for time series has attracted considerable research interest. Most of the recently used algorithms utilize the Dynamic Time Warping (DTW) distance for measuring the similarity of time series, in various areas such as science, medicine, industry, and finance. DTW is a considerably more robust distance measure for time series, which allows similar shapes to match even if they are of different lengths. Unfortunately however, several serious problems are associated with the use of DTW, such as high complexity and “one to many” problems. The present study is aimed at introducing a novel technique for improving the DTW algorithm, known as Jumping Dynamic Time Warping (JDTW). It is proven that this approach improves the efficiency with lower omission factor and reduces the noise impact of query sequence.