Ghanbari, Mohammadreza (University of Melbourne), Bagloee, Saeed (Melbourne Uni), Qi, Jianzhong (The University of Melbourne), Sarvi, Majid (University of Melbourne)

Evaluating Important Nodes in Transportation Networks Using Graph Neural Networks and Communicability Centrality

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Identifying important nodes in transport networks is a critical task for optimizing infrastructure, enhancing resilience, and managing traffic flow. Traditional graph-based techniques, such as centrality measures, offer heuristic insights into node significance but often overlook node features and suffer from scalability issues. Recent advances in graph neural networks (GNNs) have demonstrated strong performance by learning from both node features and structural information, though they typically rely on supervised learning. To address these limitations, we propose TCGNN, an unsupervised incorporation of GCN and centrality-based model that integrates graph structure and node features. Our approach offers accurate and scalable node importance ranking while requiring no labeled data. The results on node classification show that our model outperforms the traditional models by an average of 4.2% in terms of accuracy. Additionally, due to approximation adaptation, it is highly suitable for large-scale transport networks.