Integrating Street Perception and Multidimensional Geo-Spatial Analytics: An Algorithm-Driven Framework for Assessing Green Exposure and Gender Equity

Building inclusive, high-density cities requires understanding vulnerable groups’ public space usage. While green exposure significantly impacts urban health, existing research frequently overlooks females’ specific needs regarding streetscape visual quality, green space structures, and daily travel experiences. To address this, the study investigates spatial disparities in Suzhou’s historic district. Utilizing multi-source data and mixed modeling strategies, including Partial Least Squares and Ordinary Least Squares (PLS-OLS) and eXtreme Gradient Boosting (XGBoost), the research analyzes how streetscape perceptions and green space characteristics affect female life satisfaction and expressed sentiment. Results indicate three main findings. (1) Streetscape visual features fundamentally drive subjective evaluations. Safe significantly enhances well-being, whereas boring and lively negatively impact life satisfaction, reflecting females’ acute sensitivity to environmental oppressiveness during daily travel. (2) Park diversity elevates expressed sentiment, while patch density positively influences life satisfaction, demonstrating the vital value of fragmented greenery for daily public space usage. (3) Boring precipitously diminishes life satisfaction after surpassing a specific threshold, while park diversity elevates expressed sentiment only after crossing a critical interval. The study establishes an integrated analytical framework linking visual perception, green space structure, emotional response, and satisfaction. These findings provide targeted strategies for enhancing inclusive urban design and optimizing green space allocation to improve streetscape safety and alleviate visual oppressiveness, thereby advancing gender social justice for vulnerable groups in historic districts.