Yumei Huang, Professor, School of Mathematics and Statistics, Lanzhou University

Zhongzhi Bai, Professor

Regularized Methods for Multivariate Time Series Segmentation

10:00-11:00 September 17(Saturday), Tencent Meeting ID: 331-973-588

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 etc. A time series can be considered as a piece-wise continuous function, it is natural to take its total variation (TV) norm as a prior information of this time series. In this talk, 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.