Traffic data modeling is one of the most critical tasks in Intelligent Transportation Systems (ITS). Accurate traffic data modeling has a wide spectrum of applications, such as alleviating traffic congestion, making better travel decisions, and improving the quality and efficiency of the transportation industry. Moreover, the increasing availability of traffic data provides a foundation for exploring data-driven models. Traffic data modeling includes, but is not limited to spatial dependency modeling, temporal dependency modeling, origin-destination (OD) modeling, customized traffic index modeling, and graph-based modeling. By appropriately modeling the characteristics of traffic data, applications such as traffic flow prediction, anomaly events or accident detection, and missing traffic data imputation can be studied.

The main challenge of traffic data modeling is the complex and nonlinear spatio-temporal correlations, including the topological connections of road networks and their highly dynamic nature. Due to the natural graph structure of traffic networks, a variety of graph-based deep learning frameworks (i.e., graph neural networks and their variants) in this field have been proposed. These graph-based deep learning models have stood out among traffic data modeling methods and achieved significant results. Although scholars have made efforts to explore traffic data modeling based on graph deep learning, it is still necessary to fully explore the benefits of these methods for ITS applications.

The aim of this Special Issue is to collate original research and surveys on transportation data modeling methods with graph neural networks, including graph representation methods, various graph neural network frameworks, and ITS applications.

Potential topics include but are not limited to the following:

• Traffic knowledge graph representation, inference and applications
• Short and long-term traffic flow prediction using graph neural networks
• Traffic data analysis such as data denoising, missing data imputation, and abnormal detection using graph neural networks
• Graph and hypergraph neural networks and their transportation applications
• Graph neural networks with knowledge embedding and their transportation applications
• Heterogeneous graph neural networks and their transportation applications
• Multi-task graph neural networks and their transportation applications
• Large-scale graph neural networks and their transportation applications