How do different types and landscape attributes of urban parks affect visitors' positive emotions?

• Published in: Landscape and urban planning Vol. 226; p. 104482
• Main Authors: Kong, Lingqiang, Liu, Zhifeng, Pan, Xinhao, Wang, Yihang, Guo, Xuan, Wu, Jianguo
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2022
• Subjects: Big data; China; Cultural services; Deep learning; Ecosystem services; Green and blue spaces; Human well-being; landscapes; people; sustainable development; Urban landscape sustainability; urban parks; China; Deep learning; Urban landscape sustainability; Ecosystem services; Big data; Human well-being; Green and blue spaces; Cultural services
• DOI: 10.1016/j.landurbplan.2022.104482
• ISSN: 0169-2046

•We quantified park visitors’ positive emotions using social media data and sentiment analysis.•Both park type and landscape attributes matter to the positive emotions of visitors.•We identified key characteristics of urban parks affecting visitors’ positive emotions.•Multi-sorce social media data and machine learning can facilitate urban sentiment studies. Improving the positive emotions of urban populations is essential for meeting the United Nations Sustainable Development Goals (SDGs) of “good health and well-being” and “sustainable cities and communities”. Urban parks generally may enhance people's positive sentiments, but little is known about explicitly linking the landscape composition and configuration of urban parks directly with visitors' sentiments based on social media data. The main objective of this study, therefore, was to identify key landscape attributes that influence this relationship in the Beijing metropolitan region. We first crawled 55,441 valid text data items from Sina Weibo for 99 urban parks within the fifth ring road of Beijing. Then, we quantified the positive emotions of visitors to urban parks using social media data and sentiment analysis. Finally, we evaluated the differences in visitors' positive emotions among different types of parks and used Random Forest to identify urban park attributes that were correlated with positive emotions. We found that visitors to different types of urban parks had different levels of positive emotions. Specifically, visitors to comprehensive parks and cultural relics parks were significantly happier than visitors to community parks. Visitors to parks between the third and fourth ring roads in Beijing had the lowest levels of positive emotions. Positive emotions were found to be positively correlated with park size and the mean size of water bodies but negatively correlated with large areas of impervious surfaces. This study sheds new light on the relationship between park landscape patterns and visitors' positive emotions through a new approach based on social media data. The research methods and findings may inspire similar studies in other cities and countries, which are needed to improve park planning and management and thus enhance urban sustainability.