Scholarlink

<p>Resilience has a problem: researchers and practitioners cannot agree on what resilience is. While attempts at universal definitions continue, the diversity of disciplinary perspectives has made convergence elusive. This study applies a data-centric approach to understanding resilience across disciplines using statistical modeling and machine learning techniques. We analyzed 102 unique definitions of resilience from 44 interdisciplinary papers across cybersecurity, psychology, community studies, disaster relief, ecology, network science, organizational studies, international relations, sociology, and medicine. Through word cloud analysis, word co-occurrence network analysis, term frequency-inverse document frequency (tf*idf), and latent Dirichlet allocation (LDA) analysis, we extracted the core concepts of resilience that seem to transcend disciplinary boundaries. Our LDA analysis identified 11 distinct topical dimensions that collectively describe resilience, potentially providing a more nuanced and comprehensive understanding than any single definition. This data-driven approach demonstrates a methodology for analyzing complex, interdisciplinary concepts and offers insights into attributes associated with resilient systems that may inform theoretical development, practical decision-making, and resource allocation across domains, including social–ecological systems.</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

Quantifying resilience across disciplines: a machine learning approach to analyzing resilience literature

  • 期刊论文

  • 发表日期：

    2026年3月7日

查看全文
  • Ryan Hilger

摘要

Resilience has a problem: researchers and practitioners cannot agree on what resilience is. While attempts at universal definitions continue, the diversity of disciplinary perspectives has made convergence elusive. This study applies a data-centric approach to understanding resilience across disciplines using statistical modeling and machine learning techniques. We analyzed 102 unique definitions of resilience from 44 interdisciplinary papers across cybersecurity, psychology, community studies, disaster relief, ecology, network science, organizational studies, international relations, sociology, and medicine. Through word cloud analysis, word co-occurrence network analysis, term frequency-inverse document frequency (tf*idf), and latent Dirichlet allocation (LDA) analysis, we extracted the core concepts of resilience that seem to transcend disciplinary boundaries. Our LDA analysis identified 11 distinct topical dimensions that collectively describe resilience, potentially providing a more nuanced and comprehensive understanding than any single definition. This data-driven approach demonstrates a methodology for analyzing complex, interdisciplinary concepts and offers insights into attributes associated with resilient systems that may inform theoretical development, practical decision-making, and resource allocation across domains, including social–ecological systems.