Decoding the cultural heritage tourism landscape and visitor crowding behavior from the multidimensional embodied perspective: Insights from Chinese classical gardens

Abstract

This study integrates mobile sensing technologies into a "space-perception-behavior" framework, grounded in multidimensional embodied cognition, to evaluate the spatial quality of cultural heritage tourism in classical Suzhou gardens, China. "Space" level, machine learning technology is employed for the image semantic segmentation of panoramic images to measure spatial elements. "Perception" level, text semantic segmentation analysis of various perceptions of the physical environment using social media data is conducted with SnowNLP. "Behavior" level, spatial density analysis of visitors' onsite crowd aggregation characteristics is conducted with panoramic images and online check-in data. The results reveal that the sky view index and window-to-wall ratios, along with physical microclimatic environmental factors such as temperature and humidity, significantly influence the visitors’ crowding behavior. This research contributes to developing a synergistic system to evaluate embodied cognition and spatial quality, clarifies their mutual influencing mechanisms, and provides theoretical and practical guidance for sustainable heritage tourism development.