The Recommendation Method for Hotel Selection Under Traveller Preference Characteristics: A Cloud-Based Multi-Criteria Group Decision Support Model

Abstract

The research provides a multi-criteria group decision support model for hotel selection according to traveller preference characteristics. Specifically, the study attempts to identify characteristic preferences among travellers through online reviews and then generate recommendation lists under these characteristics. Utilizing a sample of 48,756 online reviews on TripAdvisor.com, the research exploits natural language processing to acquire high-frequency feature words and employs these features to group online reviews for filtering unrelated ones. Subsequently, online reviews and fuzzy theory are combined to develop a decision support model, by utilising the information of user ratings and text comments. In addition, the related extension of the theoretic knowledge is presented. The findings demonstrate that, hotel quality varies under different characteristics and a comprehensive evaluation of hotels would encounter difficulty meeting travellers’ demands, and the proposed model exhibits great performance in addressing this condition. The study can enable travellers to make a satisfactory hotel selection that matches their expectation to a high degree and also provides implications for hoteliers to understand tourist preferences and for tourism websites to optimize management strategies.