Empirical evaluation of dissimilarity measures for time-series multiscale matching

Abstract

This paper reports the results of empirical evaluation of the dissimilarity measure for time-series multiscale matching. In order to investigate fundamental characteristics of the dissimilarity measure, we performed quantitative analysis of the induced dissimilarities using simple sine wave and its variants, and compared them with dissimilarities obtained by dynamic time warping. The results showed that differences on the amplitude, phase and trends were respectively captured by the terms on rotation angle, phase and gradient, although they also showed weakness on the linearity.