Understanding the effect of tourists’ attribute-level experiences on satisfaction – a cross-cultural study leveraging deep learning

This study investigates how cultural traits play a role regarding the effect of attribute-level experiences on tourist satisfaction. Adopting Deep Learning algorithm, we proposed Attribute-Level Sentiment Analysis Model (ASAM) to extract tourists' attribute-level experiences from online reviews. Then, based on nearly 50000 online reviews collect from TripAdvisor, we empirically find that positive attribute-level experiences exert the greater influence on individualism American tourists' satisfaction, while negative attribute-level experiences affect collectivism Chinese tourists' satisfaction. In addition, we find that attribute type moderates the effect of perceived attribute experiences on overall satisfaction. Specifically, American tourists are more influenced by positive experiences with vertical attributes, while Chinese tourists are more affected by negative experiences with horizontal attributes. This research contributes to hospitality literature by enhancing the understanding of the cross-cultural factors in influencing tourist satisfaction. These findings also shed light on practices regarding improving tourist satisfaction.