I developed a machine learning model that accurately predicts loan portfolio quality for a non-banking financial company (NBFC), resulting in a significant reduction in losses. I designed and executed deep data diving initiatives to examine relationships between variables, identifying key factors that impact loan portfolio quality. Using this information, I developed a machine learning model that accurately predicts portfolio quality based on a range of variables.
Objective: The primary objective of this project was to develop a model that accurately predicts loan portfolio quality, allowing the NBFC to make informed decisions about loan approvals and minimize potential losses. Approach
To achieve this objective, I designed and executed deep data diving initiatives to examine relationships between variables, identifying key factors that impact loan portfolio quality. Using this information, I developed a machine learning model that accurately predicts portfolio quality based on a range of variables.
Repository Contents This repository contains the code used to develop the machine learning model. Additionally, I have included a README file to provide context and overview of the project.
Requirements: To run the code in this repository, you will need the following:
Python 3.0 or higher
Google Collaboratory
Scikit-learn library
Usage
To use the machine learning model, simply open the Python Notebook file included in this repository and follow the instructions provided in the code. The data used to train and test the model is also included in the repository.
Contributor This project was developed solely by me, Nivedita for Mahindra Finance. If you have any questions or feedback, please reach out to me via niveashan@gmail.com