Application of Random Forest Algorithm To Classify Credit Status of KPR Customers at Bank BTN Based on Machine Learning

Abstract

In the banking sector, this study is very suitable for determining and improving accuracy and determining credit status classification. This study aims to apply the Exploratory Data Analysis (EDA) method in supporting credit status classification at PT. Bank Tabungan Negara KCP Lubuklinggau Persero Tbk. Exploratory Data Analysis (EDA) as data exploration and Machine Learning Algorithms such as Random Forest as modeling in determining classification. The results show that the Exploratory Data Analysis (EDA) method successfully determines data patterns, while Random Forest in modeling achieves accuracy, recall, Precision, F1-Score of 100% in predicting the credit status of KPR customers. This method is expected to be useful in making decisions on more accurate credit status by the bank.