Identifying Cultural Perceptions by Nature Language Processing Method: Differences of Historical Gardens and Urban Parks in Suzhou, China
Chang Su,Yixiu Chen,Xin Wang,Lin Song
DOI: https://doi.org/10.1145/3702329
2024-11-02
Journal on Computing and Cultural Heritage
Abstract:Historical gardens, as important components of urban green spaces, have ecological, social, and significant historical heritage value. Urban residents develop close connections with nature, culture, and historical elements through landscape perception behavior in their daily activities, leading to different preferences and value judgments. However, it is still necessary to conduct in-depth research to explore whether there are differences in cultural perception and preferences between historical gardens and newly built parks during residents' daily visits. To fill these gaps, this research utilized user-generated contents and natural language processing method, focuses on a sample of 40 historical gardens and newly built parks in Suzhou, China, to address the following three research questions: (1) Which elements do park visitors perceive more strongly in historical gardens and new parks?(2) Is there a correlation between the cultural perception preferences and the attributes of the gardens and parks?(3) What are the sentimental preferences of visitors between historical gardens and new parks? The research findings reveal that cultural and aesthetic perception is more pronounced in historical gardens compared to newly built parks, while the latter place greater emphasis on the perception of natural features. Additionally, attribute features of parks, such as area, perimeter, and public transportation facilities, equally impact visitors' landscape perception. This study distinguished the cultural perception and preference characteristics of urban residents in historical gardens and new parks, providing urban managers and urban green space planning and design professionals with more experience in areas such as the preservation and renovation of urban historical landscapes.
computer science, interdisciplinary applications