International Research Journal of Education and Technology

Data mining is essential in numerous fields, particularly in the banking industry, which presents unique challenges due to the extensive volume and complex nature of financial data. The growing customer base in banking has complicated the processes of loan approvals and credit risk evaluations. Although current methodologies provide solutions for predicting credit risk, they frequently encounter difficulties when dealing with large datasets and evolving attributes. This paper introduces a new decision support system, Ek-EDT (Extended K-Means and Enhanced Decision Tree), designed to effectively manage customer profiles and forecast loan repayment risks. By utilizing clustering and classification methods, the system evaluates loan applications and classifies applicants into various risk categories, thereby enabling banks to make well-informed lending choices. The proposed method enhances the speed of decision-making, refines feature selection, and streamlines loan approval procedures, ultimately leading to improved risk management within the banking sector.