Nonlinear Associations Between Multidimensional Environmental Factors and Urban Park Use Patterns Using Interpretable Machine Learning

Code and data for “Nonlinear Associations Between Multidimensional Environmental Factors and Urban Park Use Patterns Using Interpretable Machine Learning” This repository contains GPS trajectory data, processed stay-event records, park-level environmental indicators, and analysis code associated with the manuscript Nonlinear Associations Between Multidimensional Environmental Factors and Urban Park Use Patterns Using Interpretable Machine Learning. The study examines urban park use patterns and their associations with multidimensional environmental factors in Guangzhou, China. The dataset covers 397 urban parks and includes GPS trajectory data collected from 2021 to 2023, together with 244,974 valid park-matched stay events derived from trajectory processing. The repository also provides park-level environmental indicators and park-use-pattern information. The repository is provided to improve the transparency and reproducibility of the analyses reported in the manuscript. The trajectory data, derived stay-event records, park-level datasets, and corresponding computational scripts are provided to support verification and further research. Associated manuscript: Mai, S., Yang, M., & Ma, H. Nonlinear Associations Between Multidimensional Environmental Factors and Urban Park Use Patterns Using Interpretable Machine Learning. Study area: Guangzhou, China Study period: 2021–2023 Number of parks: 397 Data types: GPS trajectory data, valid stay events, park-level environmental indicators, and park-use patterns Valid park-matched stay events: 244,974 GPS trajectory data: Included in the repository