<![CDATA[The risk of non-performing loan is a significant issue in the financial industry, including banks and cooperatives. Loan default risks can occur due to various reasons, and one of them is the negligence of staff or subjective decision-making in loan approval. The proposed solution is to enhance an objective and accurate loan approval decision-making system through the application of machine learning technology, aiming to reduce the risk of loan default. The Random Forest algorithm has proven to be the best in predicting loan approval compared to other supervised learning models. Optimization was performed on the Random Forest algorithm through hyperparameter tuning and data balancing using SMOTE. The best accuracy obtained from several experiments was 86.2%. By implementing optimizations on the Random Forest algorithm, it is expected that the model can make loan approval predictions more objectively and accurately, serving as a reference for future loan approval system development.]]>
Copyrights © 2024
