Research on short time traffic flow forecasting method
Mei Xu · 2008
Traffic flow forecasting is a popular research topic of intelligent transportation systems.How to build suitable forecasting model and select suitable forecasting variable is a key problem.This paper took use of correlation analysis to select the close relate forecasting variables and input the variables to nonlinear regression SVM.The SVM was used to forecast traffic flow through training with history traffic flow data.Illustrated the method's efficiency through analyzing the practical traffic data.