This study aims to investigate the hotel selection differences among different types of travellers through online hotel reviews. Specifically, the study performs a detailed examination of the differences in hotel key factors, criterion importance and selection results among five types of travellers, namely, business, couples, families, friends and solo. Using a sample of 194,885 online reviews on TripAdvisor.com, this study identifies the hotel key factors and criterion importance by employing the term frequency-inverse document frequency algorithm and Word2Vec algorithm. Additionally, a bounded rationality behavioural decision support model with picture fuzzy information is proposed to address hotel selection problems for different traveller types. Our results suggest that different types of travellers present differences in hotel key factors, criterion importance and selection results. However, families and friends have similar hotel selection results. This study can serve as a reference for hotel managers in understanding traveller preferences and for tourism website optimisation.