Using fuzzy rules for prediction in tourist industry with uncertainty
Sio Iong Ao · 2004
The previous studies [V. Cho (2003)], [R. Law et al., (1999)], [J. Du Preeez et al. (2002)] have compared the prediction results from different methods. Relatively recently, neural network is introduced into the tourist forecasting field and it is found to be superior to other methods. In their study, the most recent historical value of the arrival number is used for the prediction, serving as the feeding data for the neural network [N.K. Kasabov (1997)]. Prediction of tourist numbers is important for various reasons. Hotels, restaurants and ground transportation companies, as well as the airline corporations are a few examples that require as accurate prediction as possible. Here, it is studied whether the selecting of the feeding data for the fuzzy rules generation can be done automatically. The method employed here for this purpose is hybrid econometric and fuzzy system in the loose hybrid form. Basing on the traditional econometric AR method, it is attempted to employ nonparameter method fuzzy for the prediction, as such relationship is highly nonlinear and dynamics