The Research on the Impact of Street Space Perception on Children's Friendly Cities and Residents' Emotions: An Empirical Analysis Based on Fuzhou.

ObjectiveThis study investigates the complex relationship between street space perception in child-friendly cities and residents' emotions, focusing on Fuzhou’s urban area. The aim is to explore how specific street design elements impact residents' emotional responses, with an emphasis on creating a more child-friendly urban environment. Given the growing attention to child-friendly cities, understanding how these urban spaces influence not only children but also the emotions of accompanying residents (especially parents) is essential for better urban planning. Methods The study employs a mixed-methods approach using multi-source data, including Baidu Street View images, user-generated content (UGC) from social media platforms, and official complaint data from government services. Geographic analysis technology and machine learning methods, such as convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and XGBoost with SHAP (Shapley Additive Explanations), are utilized to analyze the relationship between urban street space perception and emotional responses. Key environmental indicators like traffic flow, visual complexity, safety perception, sidewalk quality, and fence ratio are extracted from street view images using semantic segmentation and human-machine adversarial scoring. Emotional analysis of the collected textual data from social media and complaint platforms is performed through sentiment analysis models to quantify the residents' emotional responses (positive, neutral, negative) toward the urban environment. Results The results indicate that several key urban design factors significantly impact residents' emotions. Traffic flow (0.282), visual complexity (0.212), safety perception (0.21), and fence ratio (0.183) are identified as the most influential indicators. For example, high traffic flow tends to evoke negative emotions due to safety concerns for children, while an optimized visual complexity stimulates positive emotional responses. Areas like Guangming Port, which underwent child-friendly renovations, were found to have low emotional feedback, indicating the need for both design and residents' practical usage to align for effective urban space development. Conversely, streets like Pu Shang Avenue, which have not undergone specific child-friendly modifications, showed positive emotional responses, highlighting the importance of balancing urban functionality with child-friendliness.SHAP analysis further reveals the nonlinear and complex effects of various factors on emotional responses. Traffic flow, visual complexity, safety perception, and fence ratio are critical, while other factors like pavement quality and openness contribute moderately. Visual complexity was found to positively affect emotional responses when balanced, but excessive complexity negatively impacted residents’ feelings. Additionally, safety features like pedestrian crossing signals and fences enhance emotional well-being, especially in areas with high foot traffic. Conclusion The study concludes that optimizing street space perception in child-friendly cities requires a comprehensive approach. Key suggestions include regulating traffic flow, enhancing safety measures, and optimizing visual complexity to foster positive emotional responses among residents. Urban planners must ensure that child-friendly city designs not only meet children's physical needs but also consider the emotional and practical experiences of parents and other residents. Child-friendly projects must emphasize both functionality and emotional well-being by improving public safety, fostering engaging visual environments, and creating balanced spaces that serve diverse needs.The research provides theoretical insights and practical recommendations for the ongoing development of child-friendly cities in China, emphasizing that successful urban designs are those that integrate emotional feedback from residents. Furthermore, the use of advanced machine learning models like XGBoost and SHAP allows urban planners to more accurately predict the emotional impacts of specific urban features, providing a valuable tool for future city planning efforts. The study highlights the necessity of involving residents' emotions in the design process to ensure a holistic approach to child-friendly urban development.