The application of STGCN with integrated attention mechanism in traffic speed prediction

Traffic management relies on accurate data prediction, and the rapid development of artificial intelligence technologies provides new technical means for prediction methods. This paper proposes a spatio-temporal graph convolutional neural network (ASTGCN) model integrated with an attention mechanism. The model captures the spatio-temporal dynamic correlation of traffic flow through the spatio-temporal attention mechanism. By using multiple ST-blocks, it performs graph convolution and temporal convolution operations on the data, and ultimately outputs traffic flow predictions through fully connected layers. We conduct experiments on the PeMSD4 dataset, and the results demonstrate that the model outperforms baseline models in terms of performance.