SSDTW: Shape segment dynamic time warping
作者：
Highlights：
• We propose a novel DTW algorithm called shape segment dynamic time warping (SSDTW).
• SSDTW improves DTW by considering the structural information of time series data.
• SSDTW reflects the peripheral information in time series data by utilizing MODWT.
• SSDTW finds more accurate alignment paths than DTW, DDTW, and shapeDTW.
• We utilized ground truth alignments as benchmarks for various time series data.
摘要
•We propose a novel DTW algorithm called shape segment dynamic time warping (SSDTW).•SSDTW improves DTW by considering the structural information of time series data.•SSDTW reflects the peripheral information in time series data by utilizing MODWT.•SSDTW finds more accurate alignment paths than DTW, DDTW, and shapeDTW.•We utilized ground truth alignments as benchmarks for various time series data.
论文关键词：Time series data,Dynamic time warping (DTW),Shape segment dynamic time warping (SSDTW),Maximal overlap discrete wavelet transform (MODWT),Alignment path
论文评审过程：Received 14 September 2018, Revised 7 January 2020, Accepted 5 February 2020, Available online 6 February 2020, Version of Record 19 February 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113291
