The constraints of road network topology and dynamically changing traffic states over time make the task of traffic flow prediction extremely challenging. Most existing methods use CNNs or GCNs to capture spatial correlation. However, convolution operator-based methods are far from optimal in their ability to fuse node features and topology to adequately model spatial correlation. In order to model the spatio-temporal features of traffic flow more effectively, this paper proposes a traffic flow prediction model, the Spatio-Temporal Graph Attention Network (STGAN), which is based on graph attention mechanisms and residually connected gated recurrent units. Specifically, a graph attention mechanism and a random wandering mechanism are used to extract spatial features of the traffic network, and gated recurrent units with residual connections are used to extract temporal features. Experimental results on real-world public transportation datasets show that our approach not only yields state-of-the-art performance, but also exhibits competitive computational efficiency and improves the accuracy of traffic flow prediction.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Spatio-temporal graph attention networks for traffic prediction


    Weitere Titelangaben:

    C. MA ET AL.
    TRANSPORTATION LETTERS


    Beteiligte:
    Ma, Chuang (Autor:in) / Yan, Li (Autor:in) / Xu, Guangxia (Autor:in)

    Erschienen in:

    Transportation Letters ; 16 , 9 ; 978-988


    Erscheinungsdatum :

    20.10.2024


    Format / Umfang :

    11 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Spatio-Temporal Graph Attention Convolution Network for Traffic Flow Forecasting

    Liu, Kun / Zhu, Yifan / Wang, Xiao et al. | Transportation Research Record | 2024


    Spatio-temporal Dynamic Graph Convolutional Probability Sparse Attention Networks for Traffic Flow Forecasting

    Chen, Linlong / Chen, Linbiao / Wang, Hongyan et al. | Springer Verlag | 2025


    Traffic Flow Prediction Based on Spatio-Temporal Aggregated Graph Neural Networks

    Wu, Shuangshuang / Hu, Yao | Transportation Research Record | 2025


    Self-Attention Graph Convolution Imputation Network for Spatio-Temporal Traffic Data

    Wei, Xiulan / Zhang, Yong / Wang, Shaofan et al. | IEEE | 2024


    Automated Dilated Spatio-Temporal Synchronous Graph Modeling for Traffic Prediction

    Jin, Guangyin / Li, Fuxian / Zhang, Jinlei et al. | IEEE | 2023