Loan Scope: Predicting Real Time CIBIL Impact and Recovery

Abstract

This project aims to develop an intelligent system to predict an individual's future CIBIL score using machine learning models. It leverages a Random Forest Regressor trained on historical financial data, including variables such as current CIBIL scores, loan amounts, credit card usage, and repayment histories, to accurately predict changes in credit behavior. The model dynamically analyzes how new loans, late payments, and other credit activities influence the future score, making it adaptable to a wide range of user profiles. The system's architecture includes a Flask-based backend that handles user registration, login, and the fetching of user financial data based on PAN numbers from a pre-existing database. By collecting user data through a CSV file, the model uses this input to generate a prediction for future CIBIL scores. It adjusts predictions based on factors such as new loans, penalties for late payments, and rewards for timely payments, providing users with insights into how their actions could impact their financial standing. The project is designed to offer real-time insights and accurate forecasts oncredit scores, empowering users to manage their credit health proactively. This machine learning-driven approach enhances the traditional CIBIL score analysis with a data driven framework