AI-Driven Credit Scoring and Credit Line Solution for the Unreserved and Self-Employed

The financial industry's rapid evolution emphasizes the importance of accurate credit scoring, crucial for decision-making and timely access to loans. Challenges arise due to irregular income streams experienced by self-employed individuals, leading to uncertainty in demonstrating consistent repayment capacity to lenders. Self-employed individuals are often perceived as higher-risk borrowers due to income instability, resulting in stricter lending criteria, higher interest rates, or reduced loan amounts. Customer credit classes are based on default risk, with precise risk assessment vital for profitability. The proposed system employs a Random Forest Classifier Ensemble Learning model on a Two-wheeler Motorcycle Indian Dataset to enhance credit risk assessment comprehensively, especially for self-employed individuals. Credit line predictions incorporating ensemble learning on random forest are being utilized to provide loan amount estimations along with risk considerations, either by taking into account bank statements or other relevant financial data. The proposed system vouches to deliver an accurate result for those who seek aid from financial institutions. The proposed model achieves an accuracy of 87% and good MSE score of 25680.607346699337 by using the Random forest and Ensemble Learning model. The explainable AI graph serves as a valuable tool for transparency and understanding in AI systems, helping users trust and interpret the model's decisions more effectively.