On the basis ofTEI@Imethodology’s theoretical framework, a TEI@Ibased foreign exchange rates forecasting model is proposed, in which econometrical models are used to forecast the main trends of the rates, the nonlinear components of the rates are analyzed by using artificial neural network (ANN) models and the impacts of irregular and the infrequent future factors on the rates are explored using text mining and rulebased expert systems techniques. A fully novel nonlinear integrated forecasting approach with error correction and judgmental adjustment is formulated by means of support vector regression technique. For further illustration, the effectiveness of the TEI@Ibased foreign exchange rates forecasting model was verified by the three foreign exchange rates.