Higher Accuracy on Loan Eligibility Prediction using Random Forest Algorithm over Decision Tree Algorithm

Abstract

Aim: The main goal of this research is to create a proficient prediction of Supervised Machine learning algorithms for checking whether a person is eligible to get loan approval or not. Material and Methods: Random forest algorithm and Decision Tree algorithm are the two groups of algorithms that are applied in this study. The paper consists of around 981 rows and 13 columns used to train and test the Machine learning models to verify the loan status of persons. The experimental research had performed with N=10 iterations for each algorithm by taking a G-power of 80%. Results: The outcomes of this experimental research mean accuracy is over 89.94% in the Random forest algorithm and 86.69% in the Decision Tree algorithm. After performing the statistical analysis, independent sample tests show that the significant difference between the two algorithms is p = 0.024 where p < 0.05. It shows that the Random forest algorithm and Decision Tree algorithm are more stable. Conclusion: This research work aims to implement the innovative approach of skillful Machine learning algorithms for innovative loan eligibility prediction and also to improve accuracy in existing algorithms like the Random Forest algorithm and Decision Tree algorithm. By comparing all the analyses of experimental results. It is clearly shown that the Random forest algorithm has the highest accuracy over the Decision Tree algorithm for loan eligibility prediction.