A fast and accurate similarity measure for long time series classification based on local extrema and dynamic time warping

作者：

Highlights：

• A new parameter-free measure is proposed for classification of long time series.

• The newly proposed ‘LE-DTW’ is based on local-extrema and dynamic time warping.

• Experiments on real-world datasets show efficiency and accuracy of our proposal against popular distance-based methods.

摘要

•A new parameter-free measure is proposed for classification of long time series.•The newly proposed ‘LE-DTW’ is based on local-extrema and dynamic time warping.•Experiments on real-world datasets show efficiency and accuracy of our proposal against popular distance-based methods.

论文关键词：Classification,DTW,Local extrema,Long time series,Similarity measures,Time series

论文评审过程：Received 17 May 2020, Revised 7 September 2020, Accepted 24 November 2020, Available online 3 December 2020, Version of Record 11 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114374