DynaKey-GNN: An efficient dynamic key-node multi-graph neural network for spatio-temporal traffic flow forecasting (SCI 中科院Q1 Top)
Published in Engineering Applications of Artificial Intelligence, 2025
A dynamic key-node multi-graph neural network for traffic flow prediction that achieves up to 12.37% higher accuracy than state-of-the-art baselines.
Recommended citation: Fan, G., Sabri, A. Q. M., Rahman, S. S. A., & Pan, L. (2025). DynaKey-GNN: An efficient dynamic key-node multi-graph neural network for spatio-temporal traffic flow forecasting. Engineering Applications of Artificial Intelligence, 159, 111757.