Scholarlink

<p>South Korea’s rapid ageing and declining birth rates challenge urban planning, yet evidence on how demographic shifts reshape commercial districts remains limited. This study measures social and economic vitality across Seoul’s neighbourhood commercial districts in 2022 and projects vitality to 2030 using explainable machine-learning models. Five datasets covering commercial-district boundaries, mobile-based floating population, sales, store counts, and registered resident population projections are combined to estimate age-specific and regional differences between Gangnam and Gangbuk. Results indicate overall declines in vitality but heterogeneous patterns, including districts where vitality is expected to increase despite demographic headwinds. SHAP-based interpretation suggests that the most influential predictors of sales in Gangnam are Food, Retail, and GDP, whereas Gangbuk exhibits a more diversified set of influential factors, including Living Services and Academics/Education. These findings provide an interpretable forecasting framework to support place-specific policy responses to demographic transition.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Forecasting social and economic vitality in Seoul’s neighbourhood commercial districts under demographic change using explainable XGBoost models

  • 期刊论文

  • 发表日期：

    2026年4月7日

查看全文
  • Juwon Hwang,
  • Seongman Jang

摘要

South Korea’s rapid ageing and declining birth rates challenge urban planning, yet evidence on how demographic shifts reshape commercial districts remains limited. This study measures social and economic vitality across Seoul’s neighbourhood commercial districts in 2022 and projects vitality to 2030 using explainable machine-learning models. Five datasets covering commercial-district boundaries, mobile-based floating population, sales, store counts, and registered resident population projections are combined to estimate age-specific and regional differences between Gangnam and Gangbuk. Results indicate overall declines in vitality but heterogeneous patterns, including districts where vitality is expected to increase despite demographic headwinds. SHAP-based interpretation suggests that the most influential predictors of sales in Gangnam are Food, Retail, and GDP, whereas Gangbuk exhibits a more diversified set of influential factors, including Living Services and Academics/Education. These findings provide an interpretable forecasting framework to support place-specific policy responses to demographic transition.