Improve Accuracy in Prediction of Credit Card Approval using a Novel Xgboost compared with Decision Tree Algorithm

Abstract

The purpose of this investigation is to evaluate the Xgboost Classifier in contrast to the Decision Tree in order to make an accurate forecast regarding credit card approval (DT). The approval of a credit card is accomplished through the use of Xgboost Classifier with a number of samples equal to ten (N = ten), as well as Decision Tree (N = ten). The credit card approval dataset includes 19 characteristics that are taken into consideration when deciding whether or not to grant a person a credit card. According to the statistical analysis performed by SPSS, the accuracy of the Xgboost Classifier is 87.96%, which appears to be superior to the accuracy of the Decision Tree, which is 79.38% correspondingly, with a significant difference of p=0.001 (p0.05, 2-tailed). According to the findings, the accuracy of the Xgboost Classifier appears to greatly outperform that of the Decision Tree (DT) when it comes to predicting whether or not a credit card application will be approved.