The financial sector operates in a multifaceted and ever-changing environment marked by diverse risk factors, with credit risk being particularly noteworthy. This research paper introduces an innovative method for evaluating credit risk, utilizing fuzzy sets to adeptly manage the inherent uncertainties in the process. By incorporating demographic variables such as age, Employer LGA, income, marital status, education status, and bank statement variables like average credit, opening balance, average debt, and closing balance, the proposed system assigns a credit risk category to loan applicants. Employing a fuzzy inference system that applies fuzzy logic to assess the creditworthiness of applicants based on input variables, the system was assessed for practicality and accuracy using a dataset of historical loan applications. The findings revealed that the proposed system achieved a commendable accuracy rate of 72%, demonstrating its efficacy in predicting credit risk. The implications of this research are substantial, as the proposed credit risk assessment system can be implemented by banks and financial institutions to automate and improve the loan approval process. Through the adoption of this system, financial institutions can enhance the efficiency of credit risk assessment, resulting in improved profitability and more informed decision-making.
Keywords: Credit Risk, Model, Fuzzy Logic, Inference System, creditworthiness, transaction, demographic