Abstract
ABSTRACTTravelling is an activity for relaxation and self-recovery. A picture showing a place with positive emotion, e.g. amusement or contentment, holds higher power than one with negative emotion. In the era of social media, how people emotionally perceive the visual content of shared photographs was rarely interpreted. In this study, we innovatively applied two deep learning models in computer vision for the visual content and sentiment analysis of 39,117 tourists’ photographs taken in Beijing. As a result, we discovered the specific correlation between the main publisher's cognitive concepts and the eight viewer affective concepts of tourists’ photographs. Practically, we established an approach for DMOs to select a positive picture with fit content for a destination. Theoretically, this study extends the role of user-generated photographs for generating the destination image.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
