Perception study of urban green spaces in Singapore urban parks: Spatio-temporal evaluation and the relationship with land cover

Abstract

In the current era of increasing urbanization, urban green spaces play a crucial role in enhancing human well-being. However, quantifying public perceptions from text data at spatio-temporal scales remains challenging, and the relationship between urban green space perception and spatial-physical attributes requires further exploration. This study systematically examines public perceptions of urban green spaces within Singapore's urban parks from 2018 to 2022. Utilizing Twitter data, it applies large language models to conduct textual content analysis related to urban green space. The findings reveal a positive trend, with individuals expressing favorable perceptions and satisfaction towards urban green spaces in Singapore. Specifically, this study demonstrates that people's perceptions of urban green spaces are influenced by vegetation density. Higher vegetation density heightens people's awareness of spatial presence, while shrub and grassland may lead to neglect of urban green spaces as individuals focus more on themselves. Additionally, due to the spatial heterogeneity of the area, there is no clear correlation between all land covers and public satisfaction with urban green spaces in Singapore. The results also indicate a significant decrease in public perception in 2020, followed by a subsequent recovery. This fluctuation is attributed to the substantial impact of the COVID-19 pandemic, suggesting that external socio-political, economic, and public health events can impact public green space needs and spatial perceptions. In conclusion, this study contributes to the understanding of urban green spaces by effectively analyzing textual content extracted from social media data using large language models. The insights gained contribute valuable to the following discussions regarding the planning and design of urban green spaces and urban parks.