Data and Code for: Coupling Coordination Degree and Driving Factors Between Urban Polycentric Spatial Structure and Park Green Space Equity

This repository contains the core dataset and custom analytical scripts supporting the research article: "Coupling Coordination Degree and Driving Factors Between Urban Polycentric Spatial Structure and Park Green Space Equity". It provides all necessary materials for editors, peer reviewers, and future researchers to verify and reproduce the statistical analyses and machine learning visualizations presented in the study.Contents of the Repository:database.xlsx: The main panel dataset containing all evaluated spatial and socioeconomic variables for the 102 sample cities in China (including the Polycentricity Index, Accessibility Index, Gini coefficient, Coupling Coordination Degree, etc.).PDP.py & shap.py: Python scripts used for fitting the Generalized Additive Model (GAM) to analyze non-linear driving factors, calculating SHAP values, and generating Partial Dependence Plots (PDP) and SHAP feature value density distributions.ReadMe.md: Detailed documentation providing a comprehensive data dictionary and explaining the methodological protocols (such as the GUI-based GIS workflows used for spatial metric calculations).Note on Raw Spatial Data: The underlying raw spatial datasets—such as the urban park boundaries (OpenStreetMap), the 100-meter population grid from China's 7th Census, terrain relief data, and administrative boundary vectors (NGCC)—are derived from publicly accessible databases. Due to file size constraints and licensing, they are not duplicated in this repository. However, stable URLs and DOIs for accessing these raw data sources are clearly listed in the attached ReadMe file and the manuscript's Data Availability Statement.