arrow
Volume 15, Issue 2
A Total Variation Based Method for Multivariate Time Series Segmentation

Min Li, Yumei Huang & Youwei Wen

Adv. Appl. Math. Mech., 15 (2023), pp. 300-321.

Published online: 2022-12

Export citation
  • Abstract

Multivariate time series segmentation is an important problem in data mining and it has arisen in more and more practical applications in recent years. The task of time series segmentation is to partition a time series into segments by detecting the abrupt changes or anomalies in the time series. Multivariate time series segmentation can provide meaningful information for further data analysis, prediction and policy decision. A time series can be considered as a piecewise continuous function, it is natural to take its total variation norm as a prior information of this time series. In this paper, by minimizing the negative log-likelihood function of a time series, we propose a total variation based model for multivariate time series segmentation. An iterative process is applied to solve the proposed model and a search combined the dynamic programming method is designed to determine the breakpoints. The experimental results show that the proposed method is efficient for multivariate time series segmentation and it is competitive to the existing methods for multivariate time series segmentation.

  • AMS Subject Headings

65C20, 62M10, 91B84

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{AAMM-15-300, author = {Li , MinHuang , Yumei and Wen , Youwei}, title = {A Total Variation Based Method for Multivariate Time Series Segmentation}, journal = {Advances in Applied Mathematics and Mechanics}, year = {2022}, volume = {15}, number = {2}, pages = {300--321}, abstract = {

Multivariate time series segmentation is an important problem in data mining and it has arisen in more and more practical applications in recent years. The task of time series segmentation is to partition a time series into segments by detecting the abrupt changes or anomalies in the time series. Multivariate time series segmentation can provide meaningful information for further data analysis, prediction and policy decision. A time series can be considered as a piecewise continuous function, it is natural to take its total variation norm as a prior information of this time series. In this paper, by minimizing the negative log-likelihood function of a time series, we propose a total variation based model for multivariate time series segmentation. An iterative process is applied to solve the proposed model and a search combined the dynamic programming method is designed to determine the breakpoints. The experimental results show that the proposed method is efficient for multivariate time series segmentation and it is competitive to the existing methods for multivariate time series segmentation.

}, issn = {2075-1354}, doi = {https://doi.org/10.4208/aamm.OA-2021-0209}, url = {http://global-sci.org/intro/article_detail/aamm/21270.html} }
TY - JOUR T1 - A Total Variation Based Method for Multivariate Time Series Segmentation AU - Li , Min AU - Huang , Yumei AU - Wen , Youwei JO - Advances in Applied Mathematics and Mechanics VL - 2 SP - 300 EP - 321 PY - 2022 DA - 2022/12 SN - 15 DO - http://doi.org/10.4208/aamm.OA-2021-0209 UR - https://global-sci.org/intro/article_detail/aamm/21270.html KW - Multivariate time series, segmentation, total variation, dynamic programming. AB -

Multivariate time series segmentation is an important problem in data mining and it has arisen in more and more practical applications in recent years. The task of time series segmentation is to partition a time series into segments by detecting the abrupt changes or anomalies in the time series. Multivariate time series segmentation can provide meaningful information for further data analysis, prediction and policy decision. A time series can be considered as a piecewise continuous function, it is natural to take its total variation norm as a prior information of this time series. In this paper, by minimizing the negative log-likelihood function of a time series, we propose a total variation based model for multivariate time series segmentation. An iterative process is applied to solve the proposed model and a search combined the dynamic programming method is designed to determine the breakpoints. The experimental results show that the proposed method is efficient for multivariate time series segmentation and it is competitive to the existing methods for multivariate time series segmentation.

Min Li, Yumei Huang & Youwei Wen. (2022). A Total Variation Based Method for Multivariate Time Series Segmentation. Advances in Applied Mathematics and Mechanics. 15 (2). 300-321. doi:10.4208/aamm.OA-2021-0209
Copy to clipboard
The citation has been copied to your clipboard