A Multi-Agent AI Framework for Mapping School Locations: A Case Study in Chaozhou, China

China’s rapid urbanization has created a significant misalignment with the slower, policy-oriented allocation of educational resources, yet a critical barrier to studying this spatial inequality is the lack of high-quality, fine-grained data on different schools. This study proposes a novel multi-agent AI (Artificial Intelligence) framework to automate the construction of a standardized geospatial dataset for schools, demonstrated through a case study of Chaozhou where data from diverse sources were integrated, validated, and annotated to create a verified dataset of 638 K-12 institutions. The resulting dataset enabled the estimation of school land area, analysis of morphological characteristics, and comparison of school distribution against population density, thereby providing a robust methodology and evidence base to inform data-driven policy on educational resource allocation and advance educational equity.