Sensing perceived urban stress using space syntactical and urban building density data: A machine learning-based approach

Human well-being is an essential criterion in achieving smart and sustainable cities. Given the significant influence of stress on individuals physical and mental health, various approaches have been proposed to examine the subjective experience of stress induced by the urban built environment and its effects on human well-being. Nevertheless, conducting assessments on a large scale continues to be a significant obstacle, particularly in today's context of rapid urbanization. This study utilized advancements in Machine Learning (ML) to develop a method for measuring perceived stress by analyzing urban building density, space syntactic characteristics, and visual features of the built environment. Through the utilization of ML models, a predictive approach has been developed that can capture the perceived stress levels of urban dwellers. The results are verified with public survey data, with R-2 reaching 0.698 obtained by evaluating the mean stress scores of 25 districts in Seoul city. The findings demonstrate that the proposed approach can effectively measure perceived stress, enabling urban planners to analyze the spatial pattern of perceived stress and the influence of the built environment on this perception. This work expands current approaches, which concentrate solely on parks, open spaces, or streetscapes, by developing a more comprehensive predictive model for measuring perceived stress levels in various urban areas.