Optimizing Credit Risk Evaluation for Community Banking Using Machine Learning Methods

My project was a case study centered around optimizing the credit risk evaluation process for community banks using machine learning processes. Specifically using a categorical boosting model labeled CatBoost to handle complex loan and collateral data in order to predict a credit risk score to benefit in the credit risk analysis process for community banks. This type of process can help support a vital part of the banking and loan process to help reduce the chance of lending money out to high-risk borrowers. An additional benefit is also being able to help with financial planning of loan loss reserves to support financial policies such as CECL.