International Journal of Scientific Research and Engineering Development

Accurate traffic congestion prediction remains a challenging task due to the dynamic and interconnected nature of urban transportation systems. Existing approaches, particularly graph neural network-based models, have improved prediction performance by capturing spatial and temporal dependencies; however, many of these models rely on static graph structures, exhibit high computational complexity, and struggle to adapt to evolving traffic conditions. To address these limitations, this paper proposes a Hybrid Adaptive Temporal Graph Neural Network (HAT-GNN), designed to integrate dynamic graph learning with spatial, temporal, and attention-based modeling within a unified framework. The proposed model incorporates an adaptive graph learning module to capture time-varying relationships between traffic nodes, followed by graph convolution for spatial feature extraction and a gated recurrent unit for temporal sequence modeling. An attention mechanism is further employed to emphasize influential nodes and time steps, enhancing the model’s predictive capability. The effectiveness of the proposed approach is evaluated on benchmark traffic datasets, where it demonstrates improved performance in terms of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) compared to established baseline models. The results indicate that the integration of adaptive graph learning and attention mechanisms contributes to more accurate and robust traffic prediction, particularly under dynamic conditions. The proposed framework offers a balanced approach between predictive performance and model flexibility, making it suitable for intelligent transportation applications.

Monika, Jyoti,"HAT-GNN: A Hybrid Adaptive Temporal Graph Neural Network for Traffic Congestion Prediction" International Journal of Scientific Research and Engineering Development, V9(3): Page(767-774) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.