Fig. 1
From: Spatio-temporal transformer and graph convolutional networks based traffic flow prediction
Schematic diagram of spatial and temporal correlation. This figure shows the spatiotemporal correlation where A, B, and C represent road nodes, while \(\:t,\) \(\:{t}_{1}\), and \(\:{t}_{2}\) represent three consecutive moments on the time axis. The solid lines represent spatial correlations, while the dashed lines represent temporal correlations. Spatial correlation refers to the mutual influence between different nodes at each time step, which varies over time, with darker colors in the figure indicating stronger influence. Temporal correlation refers to the fact that a node’s traffic state at a certain moment is not only closely related to its own state at previous moments but also influenced by the states of other nodes at those previous moments.
