This article presents a literature review on enterprise credit assessment using the Random Forest model, distinguishing it from general credit assessment, which includes a broader range of entities. The study highlights the limitations of traditional methods in credit risk evaluation. The primary objective of this research is to assess the technical configurations, predictive capabilities, and ethical considerations of applying Random Forest in credit assessment. Methodologically, a literature review approach guided by PRISMA principles was adopted, focusing on relevant studies published between 2018 and 2024. The findings indicate that Random Forest models enhance predictive accuracy and effectively manage high-dimensional data, outperforming traditional statistical methods. Furthermore, the study emphasizes the need for transparency and bias mitigation in automated credit scoring systems.
