关键词: 关键词: 多图融合, 多核注意力, 空间注意力, 图注意力, 深度学习

Abstract: Abstract: Abstract: Traffic prediction is a fundamental task in intelligent transportation systems, as accurate predictions can significantly improve the utilization of public resources. To address the limitations of insufficient utilization of contextual information, imbalanced graph fusion techniques, and consideration only of static spatial relationships in existing multi-graph neural network models, a traffic flow prediction model based on spatiotemporal multi-graph fusion (STGMF) was proposed. Firstly, different spatial correlations across regions were extracted by the model through the fusion of spatial graphs, semantic graphs, and spatial-semantic graphs. Spatial attention mechanisms and graph attention mechanisms were utilized to dynamically learn the importance of different graph structures for different neighbors. Then, a multi-kernel temporal attention mechanism was employed to capture both local and global temporal dependencies. Finally, a multi-layer perceptron was utilized to predict traffic flow and obtain the final prediction values. The validity of the model was verified on the New York City Taxi dataset and the New York City Bike dataset. Experimental results show that the Root Mean Square Error (RMSE) of the proposed model STGMF is reduces by 8.56%, 2.7%, and 2.2% compared with that of Spatio-Temporal Graph Convolutional Networks (STGCN), Attention based Spatial-Temporal Graph Neural Network (ASTGNN), and Meta-Graph Convolutional Recurrent Network (MegaCRN), respectively. It can be seen that the proposed model is effective in improving prediction accuracy.

Key words: hypergraph, maximal clique, set enumeration, approximation algorithm, pivot