ARTIFICIAL INTELLIGENCE - POWERED CREDIT SCORING : REDEFINING RISK MODELS IN MODERN LENDING

Credit scoring is a critical task in the financial industry, enabling lenders to assess the creditworthiness of applicants and mitigate financial risk. With the rise of Artificial Intelligence (AI), machine learning algorithms have significantly enhanced the accuracy and efficiency of credit risk prediction. This study explores the application of ensemble-based machine learning models—Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Random Forest—in developing robust credit scoring systems. These models are trained on historical credit data to classify applicants as good or bad credit risks based on various financial and demographic features. Performance evaluation is conducted using metrics such as accuracy, precision, recall, F1-score, and AUC-ROC. The results demonstrate that boosting algorithms, particularly XGBoost and LightGBM, offer superior predictive power and faster training times compared to traditional models. This research underscores the potential of AI-driven approaches in improving credit decision-making processes, thereby promoting financial inclusion and reducing default rates.