Traffic forecasting is a core element of intelligent traffic monitoring system. Approaches based on graph neural networks have been widely used in this task to effectively capture spatial and temporal dependencies of road networks. However, these approaches can not effectively define the complicated network topology. Besides, their cascade network structures have limitations in transmitting distinct features in the time and space dimensions. In this paper, we propose a Multi-adaptive Spatiotemporal-flow Graph Neural Network (MAF-GNN) for traffic speed forecasting. MAF-GNN introduces an effective Multi-adaptive Adjacency Matrices Mechanism to capture multiple latent spatial dependencies between traffic nodes. Additionally, we propose Spatiotemporal-flow Modules aiming to further enhance feature propagation in both time and space dimensions. MAF-GNN achieves better performance than other models on two real-world datasets of public traffic network, METR-LA and PeMS-Bay, demonstrating the effectiveness of the proposed approach.
Traffic speed forecasting plays an important role in intelligent traffic monitoring systems. Existing methods mostly predefine a fixed adjacency matrix to capture the spatial correlation between sensors in a traffic network. However, there are multiple hidden spatial correlations between sensors. A single fixed adjacency matrix cannot adaptively capture multiple spatial correlations. To overcome this limitation, we proposed a novel multiadaptive spatiotemporal flow graph neural network (MAF-GNN) for traffic speed forecasting. Specifically, MAF-GNN mainly consists of a multiadaptive adjacency matrix mechanism and a spatiotemporal flow mechanism. The multiadaptive adjacency matrix mechanism was proposed to adaptively capture multiple hidden spatial correlations between sensors. The spatiotemporal flow mechanism was proposed to further enhance the capture of temporal and spatial correlations. The experimental results on two real-world traffic datasets, METR-LA and PeMS-Bay, demonstrated the superiority of MAF-GNN. MAF-GNN outperformed baseline models in 1-h ahead forecasting.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.