Short-term traffic flow forecast based on parallel long short-term memory neural network

Abstract

Accurate and real-time traffic flow forecasting is critical to the control and development of intelligent transportation systems (ITS). In recent years, many methods based on deep learning have been used in the prediction of short-term traffic flows, LSTM is also used in short-term traffic flow forecasts because of its powerful ability to process timing data. As the traffic flow data has different short-term characteristics and periodic characteristics, different characteristics may interact with each other. In this paper, we propose a parallel long and short memory neural network model, which is used to obtain the short-time characteristics and periodic characteristics of traffic flow, the experimental results on the real data set show that the model can achieve better prediction results; Finally, in order to evaluate the performance of our model, we compared the prediction effect between the model and the other different models, and explore the reason of proposing parallel network.