Short-Term Road Traffic Flow Prediction Based on Multi-Dimensional Data

In order to overcome the dynamic interference problem of internal related factors in the prediction process of road traffic flow, a deep learning road traffic flow prediction model based on multi-dimensional influence factors is proposed. On the basis of analyzing the characteristics of traffic flow time and space, this model analyzes the traffic flow characteristics of urban traffic system under the relevant factors such as working day, holidays and weather, quantifies the relevant influence factors, and constructs multi-dimensional state vectors combined with road traffic flow. Simulations show that MAE of the MDLSTM prediction model based on multi-dimensional data is reduced by 5.528 and 3.018, and RMSE is reduced by 6.827 and 2.154, respectively, compared with the traditional BP neural network and LSTM models.