Fuzzy and nonlinear programming approach for optimizing the performance of ubiquitous hotel recommendation

This study proposes a fuzzy and nonlinear programming approach for ubiquitous hotel recommendation. In the proposed approach, the weights of the attributes of a hotel differ among travelers, among locations, and over time. In addition, the weights assigned by a traveler are considered uncertain, and this uncertainty is resolved by defining these weights in fuzzy values. The overall performance of a hotel is then evaluated with the fuzzy weighted average of performance levels along all attribute dimensions. Subsequently, a nonlinear programming model is formulated and solved to derive the fuzzy values of weights that tailor the recommendation results to travelers' choices. The proposed fuzzy and nonlinear programming approach was applied to a small region in the Seatwen District, Taichung City, Taiwan, and it satisfactorily explained travelers' hotel choices in a ubiquitous environment.