Multivariate time series classification with parametric derivative dynamic time warping

A Transportation $$L^p$$ Distance for Signal Analysis

“…For the multivariate time series we compare the performance of a 1NN classifier using the L 2 and $T{L}_{\lambda }^2$ distances as well as the state-of-the-art method dynamic time warping [22] and the OT distance average over each channel:…”

A dynamic time warped clustering technique for discrete event simulation-based system analysis

“…It should be noted that the approach in Górecki and Łuczak (2015) can only be applied to a continuous signal, as derivatives are involved. Using DTW as a distance function clustering technique, can be applied to reveal and visualise the internal structure of the collected data and hence the system that generated the data.…”

“…The performance was shown to be superior to the conventional methods such as hidden Markov model (HMM). Górecki and Łuczak (2015) combined two distances in their approach: the DTW distance between multivariate time series and the distance between derivatives of the time series.…”

Hierarchical clustering of time series data with parametric derivative dynamic time warping

“…It seems best, in this case, to use a combination of the two distances. Such an approach to the process of (supervised) classification has been applied in Górecki & Luczak (2013, 2014a) (for univariate time series) and in Górecki & Luczak (2015) (for multivariate time series). In this paper, we show that a similar technique can also be applied in the case of (one-dimensional) time series clustering.…”