Road Traffic Congestion Prediction Based on Random Forest and DBSCAN Combined Model

With the continuous development of the city, the traffic problem has become increasingly serious. In order to alleviate the inconvenience of people's travel caused by traffic congestion, this paper proposes a short-term traffic congestion prediction method based on the random forest algorithm. First of all, DBSCAN is used to identify the level of traffic congestion, and then random forest algorithm is used to train and predict the historical average speed and traffic flow of urban roads. Finally, the traffic congestion level is predicted based on the combination model. Using the traffic data of high-speed road in PEMs database of the United States for simulation experiment, the experimental results show that the accuracy of this method is 94.36%, which proves the excellent performance of this method.