Spatiotemporal multi-graph convolutional networks with synthetic data for traffic volume forecasting
作者：
Highlights：
• A novel deep learning model is proposed for traffic volume forecasting.
• Synthetic data is used to capture implicit variation patterns of traffic volume.
• A new spatiotemporal dependencies generative adversarial network is proposed.
• A new spatiotemporal multi-graph convolutional network is proposed.
• Experimental results verify the superior performance of the proposed model.
摘要
•A novel deep learning model is proposed for traffic volume forecasting.•Synthetic data is used to capture implicit variation patterns of traffic volume.•A new spatiotemporal dependencies generative adversarial network is proposed.•A new spatiotemporal multi-graph convolutional network is proposed.•Experimental results verify the superior performance of the proposed model.
论文关键词：Traffic volume forecasting,Synthetic data,Multi-graph convolutional network,Generative adversarial network,Traffic volume, speed, and occupancy
论文评审过程：Received 18 February 2021, Revised 3 July 2021, Accepted 26 September 2021, Available online 1 October 2021, Version of Record 6 October 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115992
