Transforming vacant rural school buildings into age-friendly facilities under China’s aging and low-fertility context: a natural language processing–based evaluation framework
Year: 2026
Type: article
Abstract: BACKGROUND: China's aging population and declining birth rates create urgent demand for elderly care infrastructure. Rural areas face dual challenges: over 100,000 vacant schools from declining enrollment and elderly care facilities meeting less than 30% of demand. This study develops and validates a comprehensive framework for assessing rural school-to-elderly care facility conversion feasibility. METHODS: An integrated computational-expert consensus framework was developed combining natural language processing with expert validation. SciBERT extracted candidate indicators from 89 academic articles, 25 policy documents, and 31 stakeholder interviews. Fifteen multidisciplinary experts validated indicators through two-round Fuzzy Delphi process using triangular fuzzy numbers. Consensus was assessed using Kendall's coefficient of concordance with 80% acceptance threshold. RESULTS: The model achieved F1-score of 0.85, precision of 0.87, and recall of 0.83. Expert validation yielded high consensus (Kendall's W=0.781, p<0.001), producing 17 validated indicators across four dimensions: building conditions, functional conversion, environmental renovation, and socioeconomic factors. Policy support emerged as most critical (weight=0.0792), followed by resource utilization and catering function (both 0.0669). All indicators achieved consensus rates of 80-100%. CONCLUSIONS: This validated framework integrates computational objectivity with expert knowledge for evaluating rural school conversion feasibility. It provides practical guidance for policymakers and developers to address rural elderly care shortages through sustainable infrastructure adaptation, applicable to aging societies globally.
Source: BMC Geriatrics
Language: English
Subfield: General Health Professions
Field: Health Professions
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,190
OpenAlex ID: W7119475005
