Eliminating Propagation Delay: Attention-Based Spatial-Temporal Fusion Graph Convolution Network for Traffic Flow Prediction

2026-07-27

AI Summary

This work addresses the limitations of existing traffic flow prediction methods, which often neglect information propagation delays among nodes and suffer from high model complexity and low efficiency. The authors propose A-STFGCN, a novel network that explicitly models and mitigates the effects of propagation delay within graph convolution for the first time. By integrating a spatio-temporal fusion module to correct delay-induced errors and incorporating a masked multi-head self-attention mechanism, the model effectively captures both short- and long-term temporal dependencies while dynamically modeling spatial relationships. Evaluated on five real-world datasets, A-STFGCN significantly outperforms eight state-of-the-art baselines, achieving high prediction accuracy alongside computational efficiency and superior data utilization.

Abstract

Predicting traffic flow is crucial to optimizing transportation systems and improving urban mobility. Many graph convolution-based models have been proposed to extract spatial-temporal features and predict traffic flow. However, most focus on spatial-temporal and semantic correlation in topological relationships. There are two primary problems to address. Firstly, the convolutional structure in the model focuses on utilizing static spatial dependencies and spatial-temporal relationships in topological structures, while neglecting the different information propagation delays between adjacent nodes in the convolution. Secondly, these methods often stack a large number of complex structures, resulting in a substantial increase in computational time during the model training phase, thereby disregarding the model's requirements for timeliness. In this paper, we propose a novel network called the Attention-Based Spatial-Temporal Fusion Graph Convolution Network (A-STFGCN). We design a spatial-temporal fusion block to extract the spatial-temporal feature correlations with propagation delay errors removed and to capture both long-term and short-term temporal characteristics of the data within a multi-head self-attention mechanism based on a mask matrix. Extensive experiments on five real-world datasets demonstrate that our method achieves the best overall performance while having good computation and data utilization efficiency compared with the eight baseline methods.

Problem

• propagation delay
• traffic flow prediction
• spatial-temporal correlation
• computational efficiency
• graph convolution

Innovation

• propagation delay elimination
• spatial-temporal fusion
• graph convolution network
• masked multi-head self-attention
• traffic flow prediction