MT-Stnets: Multi-task spatial-temporal networks for multi-scale traffic prediction

Abstract

Existing traffic prediction models mostly focus on predicting the traffic conditions on each road, which can be considered as a fine-grained prediction. In many scenarios, however, a coarse-grained prediction, such as predicting traffic flows among different urban areas covering multiple road links, is also required to help government have a better understanding on traffic conditions from the macroscopic point of view. This is especially useful in the applications of urban planning and public transportation planning. This paper for the first time studies the novel problem of predicting multi-scaled (both fine- and coarse-grained) traffic flows, and proposes a Multi-task Spatial-Temporal Networks model entitled MT-STNets to effectively address it. Specifically, given a road graph, we first construct a coarse-grained road graph based on both the topology closeness and the traffic flow similarity among the nodes (road links). Then a cross-scale spatial-temporal feature learning and fusion technique is proposed for dealing with both the fine- and coarse-grained traffic data. To make the predictions on the two scale data consistent, a structural constraint is also introduced. We conduct extensive evaluations over two real traffic datasets, and the results demonstrate the superior performance of the proposal on both fine- and coarse-grained traffic predictions.