Motif Discovery Comparison using Multivariate Rhythm Sequence Technique and Dynamic Time Warping (DTW) in Time Series Data

Motif extraction is the process of looking for recurring patterns in time series sequences. Dynamic time warping is an algorithm that finds the distance between pairs of sequences and can be used to find clusters in a set of sequences. In this paper, we review the Multivariate Rhythm Sequence Technique of motif discovery and discussed ultrafast subsequence method Dynamic Time Warping (DTW) to identify the subsequence in the discovered motif. Dynamic Time Warping distance is applied both in ECG and Video data and comparative results of two different sequences are depicted. We demonstrated our method with the most extensive set of multi-dimensional time series data and experiments are shown.