Scholarlink

<p>This study expands its scope to 120 countries worldwide, employing the Super-SBM model to calculate their Ecological Efficiency (EE). It conducts a more comprehensive and systematic examination of EE's spatiotemporal characteristics through methods such as kernel density estimation and Exploratory Spatiotemporal Data Analysis (ESTDA). Finally, it utilizes cutting-edge machine learning techniques to identify influencing factors and their interactions, thereby enhancing prediction accuracy. The findings suggest that (1) worldwide EE is falling overall. From highest to lowest, the average EE for each continent is as follows: Europe, Asia, Africa, Oceania, South America, and North America. (2) The worldwide EE exhibits obvious spatial heterogeneity, with fewer cities with high EE, primarily centered in North America, South America, Europe, and Africa; the EE of Oceania is on an upward trend; the countries with high EE in Africa are on a downward trend; the EE of South America is fluctuating; and the EE of the Asian region is relatively stable. (3) The overall tendency of aggregation is downward, and the spatial correlation of global EE is positive. Furthermore, the distribution of EE has shown unstable spatial patterns, poor spatial cohesiveness, and notable spatial and temporal jumps in the majority of countries. (4) Economic development level has a positive impact on EE, while urbanization level exhibits fluctuating effects on EE. Other influencing factors exert negative effects on EE, with economic development level and foreign trade exerting relatively greater impacts. Concurrently, economic development level shows positive correlations with industrial structure, urbanization level, and net foreign investment inflow. Foreign trade maintains a U-shaped relationship with industrial structure and economic development level. To improve EE, it is therefore essential to increase regional and international collaboration, optimize industrial structure, boost international trade, introduce foreign investment in a reasonable manner, and promote urbanization and environment protection in concert.</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

Spatiotemporal characteristics and influencing factors of ecological efficiency throughout 120 countries: a machine-learning study

  • 期刊论文

  • 发表日期：

    2026年1月19日

查看全文
  • Dalai Ma,
  • Pengli Deng,
  • Youbin Liu,
  • Guangming Yang,
  • Jin Sun,
  • Yin Yan,
  • Ruonan Chang,
  • Chao Hu,
  • Kaihua Li

摘要

This study expands its scope to 120 countries worldwide, employing the Super-SBM model to calculate their Ecological Efficiency (EE). It conducts a more comprehensive and systematic examination of EE's spatiotemporal characteristics through methods such as kernel density estimation and Exploratory Spatiotemporal Data Analysis (ESTDA). Finally, it utilizes cutting-edge machine learning techniques to identify influencing factors and their interactions, thereby enhancing prediction accuracy. The findings suggest that (1) worldwide EE is falling overall. From highest to lowest, the average EE for each continent is as follows: Europe, Asia, Africa, Oceania, South America, and North America. (2) The worldwide EE exhibits obvious spatial heterogeneity, with fewer cities with high EE, primarily centered in North America, South America, Europe, and Africa; the EE of Oceania is on an upward trend; the countries with high EE in Africa are on a downward trend; the EE of South America is fluctuating; and the EE of the Asian region is relatively stable. (3) The overall tendency of aggregation is downward, and the spatial correlation of global EE is positive. Furthermore, the distribution of EE has shown unstable spatial patterns, poor spatial cohesiveness, and notable spatial and temporal jumps in the majority of countries. (4) Economic development level has a positive impact on EE, while urbanization level exhibits fluctuating effects on EE. Other influencing factors exert negative effects on EE, with economic development level and foreign trade exerting relatively greater impacts. Concurrently, economic development level shows positive correlations with industrial structure, urbanization level, and net foreign investment inflow. Foreign trade maintains a U-shaped relationship with industrial structure and economic development level. To improve EE, it is therefore essential to increase regional and international collaboration, optimize industrial structure, boost international trade, introduce foreign investment in a reasonable manner, and promote urbanization and environment protection in concert.