Statistical analysis of urban traffic flow using deep learning

In recent years, urbanization has brought about challenges such as population growth, increased demand for traffic, and traffic congestion. To address the need for accurate traffic condition statistics, this paper proposed an improved method that combines graph convolutional network (GCN) and long short-term memory network (LSTM) model for forecasting and statistics of traffic conditions. Through the modeling and analysis of urban road conditions and traffic flow, the combination of GCN model and LSTM model enabled more precise prediction of traffic flow trends. Experiments were carried out on the actual traffic data set of Cangzhou, Hebei. The results demonstrated that the proposed method achieved high accuracy and reliability in predicting traffic flow. By using the LSTM model to improve the GCN model, it effectively adapts to changes in urban traffic conditions while providing dependable predictions.