Combining Weather Condition Data to Predict Traffic Flow: A GRU Based Deep Learning Approach

Abstract

Traffic flow prediction is an essential component of the intelligent transportation management system (ITS). This paper combines recurrent neural network and gated recurrent unit (GRU) to predict urban traffic flow considering weather conditions. Running results show that, under the review of weather influences, our method improves predictive accuracy and also decreases the prediction error rate. To our best knowledge, this is the first time that traffic flow is predicted in urban freeways in this particular way. This study examines it with respect to extensive weather influence under Gated Recurrent Unit (GRU) based deep learning framework.