A dynamic spatial–temporal deep learning framework for traffic speed prediction on large-scale road networks

作者：

Highlights：

• Novel traffic deep learning prediction model on large-scale road networks.

• Graph convolution network and attention mechanism for spatial feature analysis.

• Sequence-to-sequence architecture for temporal feature analysis.

• Result validated using real-world traffic datasets.

摘要

•Novel traffic deep learning prediction model on large-scale road networks.•Graph convolution network and attention mechanism for spatial feature analysis.•Sequence-to-sequence architecture for temporal feature analysis.•Result validated using real-world traffic datasets.

论文关键词：Intelligent transportation system,Traffic prediction,Deep learning,Large-scale road networks

论文评审过程：Received 12 March 2021, Revised 6 January 2022, Accepted 17 January 2022, Available online 7 February 2022, Version of Record 11 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116585