Adaptive Spatio-Temporal Relation Based Transformer for Traffic Flow Prediction

As network and autonomous driving technologies rapidly advance, traffic flow prediction has become a crucial area of research. It plays a significant role in optimizing urban traffic management and enhancing road safety, drawing increasing attention from researchers. As a specific form of time-series data, traffic flow data is often used in prediction tasks utilizing large language models. Recent developments in graph data and improvements in graph neural networks have led researchers to employ methods like adjacency and Laplacian matrices for addressing relational issues among distant nodes. However, most existing methods focus on enhancing prediction performance through network architectures or using adaptive matrices to capture spatiotemporal relationships, with limited exploration into the impact of input embedding. This paper introduces an innovative approach to traffic flow prediction, the ASTRformer, which emphasizes the fusion of spatial and temporal information in historical data through an adaptive spatio-temporal relation learning mechanism. This mechanism integrates feature embedding with adaptive spatial and temporal embeddings. A learnable spatio-temporal fusion network then parameterizes these embeddings, producing the input representations. Subsequently, a transformer model captures these representations to predict future traffic flows. Experimental results on six datasets demonstrate that our method effectively captures spatio-temporal dependencies, achieving state-of-the-art performance across various prediction metrics.