Short term prediction of traffic flow based on deep learning

In recent years, the problem of urban traffic congestion has become increasingly serious. Urban rail transit is an important means to alleviate the urbanization problems such as traffic congestion and environmental pollution in the process of rapid urban development. Therefore, the short-term passenger flow prediction of urban rail transit system has become an important research direction of the current urban rail transit system. In this paper, wavelet de-noising method is used to process the original passenger flow data, and the processed data is input to the LSTM model based on the new neuron structure to learn and train the passenger flow data. Compared with the traditional k-nearest neighbor algorithm and the standard LSTM model. The prediction results show that compared with k-nearest neighbor algorithm and standard LSTM model, the accuracy of this method is improved.