<![CDATA[This study aims to compare the performance of the Random Forest algorithm and Support Vector Machine (SVM) enhanced with Adaptive Boosting in determining credit eligibility. The method was conducted using a dataset from Kaggle consisting of 614 prospective borrowers, with attributes such as gender, marital status, number of dependents, education, business status, income, loan amount, loan term, credit history, and house location. The research results show that Random Forest based on Adaptive Boosting achieved an accuracy of 92.35%, precision of 90.12%, recall of 89.87%, and a misclassification error of 7.65%, whereas SVM based on Adaptive Boosting achieved an accuracy of 85.76%, precision of 82.45%, recall of 81.93%, and a misclassification error of 14.24%. These findings indicate that Random Forest outperforms SVM in predicting credit eligibility when enhanced with Adaptive Boosting. This research contributes to the development of credit risk assessment techniques using a data mining approach that can be applied in the financial industry to improve the accuracy of credit decision-making.]]>
Copyrights © 2024
