Abstract
Aiming at the problem that the traditional similarity measurement methods cannot effectively measure the similarity of the time series with the difference both in the trend and detail, this paper proposes a new time series similarity measurement method (MP-SAX) based on the morphological pattern (MP) and symbolic aggregate approximation (SAX). According to the empirical mode decomposition (EMD), the time series are decomposed and reconstructed into the trend component and the detail component. Then, the similarity of the trend component under morphological pattern coding and that of the detail component under symbolic aggregate approximation coding are respectively calculated by the longest common subsequence (LCS). Finally, the similarity of the time series is obtained by weighted aggregation of the similarity of trend component and detail component. The MP-SAX is verified by the simulation time series and the time series from UCR Time Series Classification/Clustering Homepage. The results show that the MP-SAX can effectively measure the similarity of the time series with the changes both in trend and detail.
Highlights
With the development of modern industry and information technology, massive data through various sensing devices are generated
This paper proposes a new method for similarity measurement of time series with the differences both in trend component and detail component
The morphological pattern (MP)-symbolic aggregate approximation (SAX) can comprehensively consider the influence of the trend component and detail component of time series on similarity measurement
Summary
With the development of modern industry and information technology, massive data through various sensing devices are generated. Euclidean distance is the most widely used similarity measurement method which is easy to calculate and has clear meaning [3], [4] It has been widely used in data mining tasks of time series [5]. In order to improve the computational efficiency of similarity measurement, some methods which express the time series in a simple and feature-rich manner have been proposed, such as, describe time series from the following aspects of time series: symbolization [19], change trend [20] and shape [21]. A new similarity measurement method which aggregates the similarity of the trend and detail component of time series by weighted manner is proposed.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
