Multivariate time series clustering based on complex network

作者：

Highlights：

• The constructed networks improve the accuracy of the clustering results.

• The network features reflect the information and reduce the dimension of MTS.

• The proposed method is suitable for large MTS data clustering.

• It noes not require to manually set the number of communities.

摘要

•The constructed networks improve the accuracy of the clustering results.•The network features reflect the information and reduce the dimension of MTS.•The proposed method is suitable for large MTS data clustering.•It noes not require to manually set the number of communities.

论文关键词：Multivariate time series,Data mining,Clustering analysis,Complex network

论文评审过程：Received 15 June 2020, Revised 28 October 2020, Accepted 19 February 2021, Available online 23 February 2021, Version of Record 28 February 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107919