Abstract
Benefiting from the rapid development of geospatial big data-related technologies, intelligent transportation systems (ITS) have become a part of people’s daily life. Traffic volume forecasting is one of the indispensable tasks in ITS. The spatiotemporal graph neural network has attracted attention from academic and business domains for its powerful spatiotemporal pattern capturing capability. However, the existing work focused on the overall traffic network instead of traffic nodes, and the latter can be useful in learning different patterns among nodes. Moreover, there are few works that captured fine-grained node-specific spatiotemporal feature extraction at multiple scales at the same time. To unfold the node pattern, a node embedding parameter was designed to adaptively learn nodes patterns in adjacency matrix and graph convolution layer. To address this multi-scale problem, we adopted the idea of Res2Net and designed a hierarchical temporal attention layer and hierarchical adaptive graph convolution layer. Based on the above methods, a novel model, called Temporal Residual II Graph Convolutional Network (Tres2GCN), was proposed to capture not only multi-scale spatiotemporal but also fine-grained features. Tres2GCN was validated by comparing it with 10 baseline methods using two public traffic volume datasets. The results show that our model performs good accuracy, outperforming existing methods by up to 9.4%.
Highlights
Geo-information-based Internet of Things (IoT) devices have become the infrastructure of the intelligent transportation system, which has advanced the rapid development of the intelligent transportation system
To address the aforementioned problems, we proposed a novel spatiotemporal graph neural network framework—Temporal Residual II Graph Convolution Network (TRes2GCN)—to optimize the extraction of capturing spatiotemporal features, improving the prediction accuracy
Comparing the second row with the fourth row, the results show that the Res2Net without SE-Block is more suitable for deployment in the prediction of spatiotemporal graph neural networks
Summary
Geo-information-based Internet of Things (IoT) devices have become the infrastructure of the intelligent transportation system, which has advanced the rapid development of the intelligent transportation system. High accuracy short-term traffic forecasting, as the key mission to intelligent transportation systems (ITS), can effectively help road management, congestion relief, travel planning and many other applications. The nonlinearity and complexity of traffic flow make the spatiotemporal patterns of traffic flow dynamic, variable and difficult to understand. The temporal pattern refers to the dynamically changing traffic pattern, which shows periodicity and tendency; the spatial pattern refers to the interaction between nodes in a transportation network, which shows the traffic state at a point influenced by the upstream traffic condition of the connected road. Due to its great practical value, people have been working on more accurate forecasting models from the perspective of spatiotemporal patterns. In the early traffic forecasting research, the traditional statistical traffic prediction models focus on temporal patterns. The representative models include AutoRegressive Integrated Moving Average (ARIMA) [1], Vector
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
