Bus Station Time Probability Interval Prediction Based on DNN and NKDE
The influence factors of time for the bus at the station platform analysis,considering complex drive mode is introduced into the concept of probability interval prediction station time.The station time prediction model is established through algorithm within DNN forecasting results are obtained,the statistical error of the predicted values and the real value and multiple sub divided into sections,for each segment nonparametric kernel density estimation algorithm is used to con-struct the probability interval prediction model,estimate the reorganization of the predicted results of each section under a certain confidence level of the bus station time probability interval prediction results.Using the data from the jining city bus to leave the information such as the feasibility and effectiveness of the validation method.
probabilistic interval predictionnonparametric kernel density estimationbus dwelling timesegmentation
万方数据
维普
