构建可跨类型评估公共空间质量的数据驱动框架

Public space quality assessment lacks systematic methodologies that integrate factors across diverse spatial typologies while maintaining context-specific relevance. Current approaches remain fragmented within disciplinary boundaries, limiting comprehensive evaluation and comparative analysis across different space types. This study develops a systematic, data-driven framework for assessing public space quality through the algorithmic integration of empirical research findings. Using a 7-phase methodology, we transform 1,207 quality factors extracted from 157 peer-reviewed studies into a validated hierarchical taxonomy spanning six public space typologies: urban spaces, open spaces, green spaces, parks and waterfronts, streets and squares, and public facilities. The methodology combines semantic analysis, cross-typology distribution analysis, and domain knowledge integration to address terminological variations and functional relationships across space types. The resulting framework organizes 1,029 unique quality factors across 14 main categories and 66 subcategories, identifying 278 universal factors applicable across all space types, 397 space-specific factors unique to particular typologies, and 124 cross-cutting factors serving multiple functions. Framework validation demonstrates systematic consistency in factor organization and theoretical alignment with established research on public spaces. This research provides a systematic methodology for transforming empirical public space research into practical assessment frameworks, supporting evidence-based policy development, design quality evaluation, and comparative analysis across diverse urban contexts.