Prediction of traffic accident duration based on integrated LightGBM
Aiming at the problem that the duration of urban road traffic accidents is difficult to predict due to many factors af-fecting it,an integrated prediction model combining bagging and LightGBM is proposed,and a grid search is used to find the opti-mal hyperparameters of the model.The results show that the proposed model has higher accuracy and generalization ability com-pared with a single LightGBM model and support vector machines,neural networks and gradient boosters.Feature importance indi-cates that accident location,time of day and weather related features have a greater impact on the prediction of duration.
traffic accident durationLightGBMintegrated prediction model
万方数据
