Scholarlink

<p>A robust food production system serves as the cornerstone for ensuring food security. Integrating the conceptual connotations of food production resilience, this study constructs a multi-dimensional evaluation index system encompassing resistance, recovery, and adaptive capacity. Utilizing panel data from Xinjiang spanning 2010 to 2022, the research employs an integrated methodology—including the entropy weight method, a modified gravity model, Social Network Analysis (SNA), and the XGBoost-SHAP model—to systematically analyze the resilience levels, structural network characteristics, and underlying driving mechanisms of food production in the region.The findings indicate that: From 2010 to 2022, the resilience level of food production in Xinjiang exhibited a continuous growth trend, characterized by spatial heterogeneity with a relatively narrow gap. During the study period, the network connectivity of food production resilience in Xinjiang became increasingly tight; however, it was characterized by a low network density along with a topology of high clustering and a short average path. Asymmetric features were observed between input and output regions, accompanied by a decrease in the number of spillovers across blocks. Per capita cultivated land area (X5), traffic accessibility (X15), agricultural technological progress (X12), and average annual temperature (X1) ranked as the top four influential factors, among which the interaction effect between per capita cultivated land area (X5) and traffic accessibility (X15) was the most significant. These research insights can provide valuable references for safeguarding national food security.</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

Evolution of network structure and driving mechanisms of food production resilience in arid regions: a machine learning-based approach

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月1日

查看全文
  • Yu Sun,
  • Xinwu Li,
  • Ben Tong,
  • Xiao Ding,
  • Hankun Ye

摘要

A robust food production system serves as the cornerstone for ensuring food security. Integrating the conceptual connotations of food production resilience, this study constructs a multi-dimensional evaluation index system encompassing resistance, recovery, and adaptive capacity. Utilizing panel data from Xinjiang spanning 2010 to 2022, the research employs an integrated methodology—including the entropy weight method, a modified gravity model, Social Network Analysis (SNA), and the XGBoost-SHAP model—to systematically analyze the resilience levels, structural network characteristics, and underlying driving mechanisms of food production in the region.The findings indicate that: From 2010 to 2022, the resilience level of food production in Xinjiang exhibited a continuous growth trend, characterized by spatial heterogeneity with a relatively narrow gap. During the study period, the network connectivity of food production resilience in Xinjiang became increasingly tight; however, it was characterized by a low network density along with a topology of high clustering and a short average path. Asymmetric features were observed between input and output regions, accompanied by a decrease in the number of spillovers across blocks. Per capita cultivated land area (X5), traffic accessibility (X15), agricultural technological progress (X12), and average annual temperature (X1) ranked as the top four influential factors, among which the interaction effect between per capita cultivated land area (X5) and traffic accessibility (X15) was the most significant. These research insights can provide valuable references for safeguarding national food security.