Jun 2023 - Aug 2023
• Designed a credit score model leveraging deep learning and ML methods, akin to FICO’s standard credit model, yielding accuracy scores: 79.65% via XGBoost Classifier, 79.65% via Random Forest Classifier, and 72.58% through DNN optimising performance through feature selection via chi-squared testing & feature importance analysis.
• Faircent Case Study:Researched India’s first P2P lending platform, its innovative business model connecting borrowers,lenders directly, and its credit assessment algorithm empowering both parties in the financial ecosystem.
• Investigated credit score impact on traditional lending and FinTechs’ use of alternative data to assess creditworthiness. Investigated machine learning tools in risk assessment, enabling personalized loan offers for improved experience.