Multi-scale temporal features extraction based graph convolutional network with attention for multivariate time series prediction

作者：

Highlights：

• A novel GCN model is proposed for multivariate time series prediction.

• EMD is used to extract multi-scale temporal features of original time series.

• Multi-head attention mechanism is utilized to explore the spatial dependencies.

• Real datasets from various fields confirms the superiority of the method.

摘要

•A novel GCN model is proposed for multivariate time series prediction.•EMD is used to extract multi-scale temporal features of original time series.•Multi-head attention mechanism is utilized to explore the spatial dependencies.•Real datasets from various fields confirms the superiority of the method.

论文关键词：Multivariate time series prediction,Features extraction,Multi-head attention,Graph neural network

论文评审过程：Received 23 October 2021, Revised 21 March 2022, Accepted 27 March 2022, Available online 30 March 2022, Version of Record 1 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117011