中文摘要：

绿色信贷信用风险评估过程面临复杂性、非线性以及不确定性等问题，现阶段商业银行采用的传统评估方法较难适用。为此，将组合分类前沿研究领域的随机森林算法应用到该评估过程中，在建立较为全面、综合的评估指标体系的基础上，构建了基于随机森林算法的绿色信贷信用风险评估模型，并以重污染行业上市公司为对象进行了实例分析。与传统模型评估结果的对比表明，该评估模型实现速度更快，评估准确率更高，较为有效地提升了评估效率。

英文摘要：

Green-credit risk assessment is one of the important bases to implement green-credit policy and realize the goal of green economy. However, the current methods that commercial banks take can hardly adapt to the characteristics of complexity, nonlinearity and uncertainty of the green-credit risk evaluation. Thus, the advanced random forest algorithm is applied to the green-credit risk assessment process. Based on the establishment of a more practical green-credit risk evaluation index system, a green-credit risk assessment model combined with random forest algorithm has been constructed, then use it to analyze the relevant risk of a listed companies of heavy pollution industry. With the comparison with traditional model, it shows that the green-credit risk assessment model based on random forest algorithm has faster speed, higher the whole process of assessment. classification accuracy, which turns out to improve the efficiency of