A HYBRID SYSTEM FOR CREDIT RISK EVALUATION IN NIGERIA MICRO-FINANCE BANKS (A NIRSAL MICRO-FINANCE BANK LTD LESSON)

Abstract

The Nigeria Incentive-Based Risk Sharing System for Agricultural Lending (NIRSAL Plc.) Microfinance Bank is facing financial risk due to non-performing loans, which may lead to loan losses. This is because poor decision-making procedures are employed when awarding these loans, potentially resulting in investors losing money on the borrowers' investments. To solve this issue, a hybrid system based on a belief network and fuzzy-inference model is being developed. Using inputs such as Loan_Request, Collateral_Value, Available_Cash, Average_Income, Deposit_Rate, Withdrawal_Rate, Average_Withdrawal, and Average_Deposit to produce 4 outputs: capital, character, capacity, and collateral. The fuzzy if/then rules are then utilized to assess the degree of credit risk and generate feasible suggestions. The Information used in this paper was obtained from NIRSAL Micro-Finance Bank. The work was executed with Java programming language, HTML, CSS, and MySql. The results of the experiment demonstrated that the mechanism accurately assessed the credit risk and produced the risk level needed for a potential suggestion that will minimize the risk level and the number of default loans incurred in Nirsal Micro-Finance Bank.