Deep Learning Methods in Short-Term Traffic Prediction: A Survey

Nowadays, traffic congestion has become a serious problem that plagues the development of many cities aroundthe world and travel life urban residents. Compared with costly long implementation cyclemeasures such as promotion public transportation construction, vehicle restriction, road reconstruction, etc., prediction is lowest cost best means to solve congestion. Relevant departmentscan give early warnings on congested sections based results prediction, rationalize thedistribution police forces, problem. At same time, due increasingreal-time requirements current short-term subject widespread concern research. Currently, most widely used model for are deeplearning models. This survey studied relevant literature use deep learning models shortterm in top journals recent years, summarized currentcommonly datasets, mainstream their applications this field. Finally, challenges future trends applied field discussed.