Utilizing Classifier Algorithms to Analyze Lending Activity at The Sumsel Babel Bank's Pagar Alam Branch

Abstract

Everyday living requires accurate information, and knowledge will play a significant role in civilization's present and future growth. It is not sufficient to rely solely on operational data when using existing data in information systems to assist decision-making activities; data analysis is necessary to fully realize the potential of the information already available. The government and banking currently work together to distribute foreign exchange credit, which helps MSMEs who want to expand their businesses by providing additional capital. Bad credit cannot be separated from lousy credit when granting bank credit, one of the issues that banks nowadays frequently face. Additionally, a credit analyst must conduct manual research and analysis to evaluate the business circumstances of potential debtors that are anticipated to affect their capacity to perform their obligations to the Bank while reviewing the distribution of foreign credit to MSMEs. In this research, a classifier algorithm was applied to create a prediction model to predict the customer before the lending application was used and process to pass the lending process in Bank Sumsel, branch Pagar Alam. The experiment was conducted, and based on our data and model, the result obtained 85.54% accuracy based on the Random Forest classifier model. The result shows the algorithm is entirely reasonable in predicting customer data.