Prediction of Road Traffic Congestion Based on Random Forest

TLDR

Urban modernization has increased vehicle numbers, worsening traffic congestion, and the random forest algorithm offers high robustness, performance, and practicability. The study constructs a traffic congestion prediction model using the random forest classification algorithm. The model employs random forest with inputs of weather, time period, road conditions, road quality, and holidays to forecast traffic congestion. The random forest model achieved 87.5 % prediction accuracy, low generalization error, fast computation, and strong applicability for predicting congested conditions.

Abstract

With the process of urban modernization becoming faster and faster, there are more and more vehicles in the city, and the situation of urban traffic congestion is becoming more and more serious. In this paper, a model of traffic congestion prediction is constructed by using machine learning classification algorithm - random forest to construct traffic congestion state prediction model. The random forest algorithm has the characteristics of high robustness, high performance and high practicability. The weather conditions, time period, special conditions of road, road quality and holiday are used as model input variables to establish road traffic forecasting model. Finally, the results show that the traffic prediction model established by using the random forest classification algorithm has a prediction accuracy of 87.5%, and the generalization error is low, and it can be effectively predicted. Moreover, the calculation speed is fast, and it has stronger applicability to the prediction of congested condition.