Predicting freight with fuzzy granular computing and support vector machine model

To improve the precision and reliability in predicting freight, we have proposed a forecasting model based on the fuzzy granular information and support vector (SVM) and regression machine. For the forecasting model, the freight can be considered as a nonlinear time series and the time series analysis method is adopted to predict the change in freight using SVM regression. Due to a large and nonlinear data, the granular information is adopted to divide the data into three segments, and each segment is predicted.