Scholarlink

<p>This study explores the application of a multi-layer deep neural network to identify green credit risks, with a focus on the role of anti-corruption measures. Using data from 36 Chinese banks, the model incorporates indicators of transparency and accountability to enhance risk prediction. Based on expert judgment and Analytic Hierarchy Process (AHP) analysis, the transparency and accountability indicators were weighted at 0.7 and 0.3, respectively. The model achieved strong performance, maintaining stability under data noise and outperforming traditional methods such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), eXtreme Gradient Boosting (XGBoost), and Deep Belief Network (DBN), with a recall rate of 0.84. Performance improved as the intensity of anti-corruption measures increased, suggesting that governance factors significantly enhance model reliability. These results demonstrate the potential of advanced machine learning techniques in financial risk assessment and emphasize the importance of institutional transparency in promoting sustainable green finance.</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

Green credit risk identification and anti-corruption measures under the application of the multi-layer deep network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月13日

查看全文
  • Zeyu Wang,
  • Caimeng Wang,
  • Zhili Bai,
  • Song Song

摘要

This study explores the application of a multi-layer deep neural network to identify green credit risks, with a focus on the role of anti-corruption measures. Using data from 36 Chinese banks, the model incorporates indicators of transparency and accountability to enhance risk prediction. Based on expert judgment and Analytic Hierarchy Process (AHP) analysis, the transparency and accountability indicators were weighted at 0.7 and 0.3, respectively. The model achieved strong performance, maintaining stability under data noise and outperforming traditional methods such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), eXtreme Gradient Boosting (XGBoost), and Deep Belief Network (DBN), with a recall rate of 0.84. Performance improved as the intensity of anti-corruption measures increased, suggesting that governance factors significantly enhance model reliability. These results demonstrate the potential of advanced machine learning techniques in financial risk assessment and emphasize the importance of institutional transparency in promoting sustainable green finance.