Cloud decision support model for selecting hotels on TripAdvisor.com with probabilistic linguistic information

Abstract

Hotel selection on tourism websites has been a prevailing trend in traveling in recent years, and selecting a suitable hotel is a high-risk decision for millions of tourists because of the intrinsic imperceptibility of the provided products. Decision support models for hotel selection have drawn the interest of numerous scholars. The problem is that some existing models cannot address substantial amounts of online information, and they neglect the influence of interrelationships among different criteria on tourists’ decisions. To cover these defects, an applicable hotel decision support model is developed for tourists utilizing online reviews on TripAdvisor.com. Considering a great deal of review information associated with hotels posted by numerous tourists on TripAdvisor.com, probabilistic linguistic term sets (PLTSs) are introduced to summarize this information statistically. Processing qualitative concepts requires effective support of reliable tools; then, a cloud model can be employed to deal with probabilistic linguistic information. First, PLTSs are converted, and a novel concept of probabilistic linguistic integrated cloud (PLIC) is proposed. Moreover, the essential algorithms and distance measure of PLICs are defined. Two information fusion tools are subsequently presented based on Heronian mean operator. Then, the hotel decision support model is established. Finally, a hotel selection problem on TripAdvisor.com is provided to demonstrate the model. Its stability and validity are further verified by a sensitivity analysis and comparison with other extant methods.