https://github.com/ujblockchain/credit_card_default_prediction
https://github.com/ujblockchain/credit_card_default_prediction
Last synced: about 1 year ago
JSON representation
- Host: GitHub
- URL: https://github.com/ujblockchain/credit_card_default_prediction
- Owner: ujblockchain
- License: mit
- Created: 2024-05-03T09:17:26.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-05-23T12:50:49.000Z (about 2 years ago)
- Last Synced: 2025-01-14T12:13:47.058Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 2.01 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Machine Learning: Credit Card Default Prediction
A machine learning practical application for credit loaning institutions - which can be used to reduce the loss incurred due to defaulting clients.
## Background
A default is the failure to make the required payments on a debt, this debt can be secured or unsecured. Examples of a secured debt include, the mortgage loan secured for a home or a business loan secured by a company's assets. The default occurs when the borrower can no longer make payments in a timely manner and the asset or collateral used to secure the debt is now in jeopardy.
For unsecured debts, defaults can occur on credit card balances which often leads to a reduction in the defaulter's credit score and this affects their ability to secure a future loan.
## Problem statement
A bank in Taiwan has collected data on their credit card clients between April and September 2005. The dataset contains information on default payments, demographic factors, credit data, history of payment, and bill statements. The company wants to use the data collected extract insights that will assist their risk team to reduce company losses incurred by defaulting clients.
The task is to build a machine learning model that will be used to predict the likelihood of client defaulting on their credit card payments which can be used by the risk team.
# Dataset
The dataset was collected from [Kaggle](https://www.kaggle.com/datasets/uciml/default-of-credit-card-clients-dataset/data) and contains information on default payments from credit card clients in Taiwan between April and September 2005. The dataset has 25 variables, including the target variable - default status.
# Machine Learning
For this classification problem - we have used a logistic regression model and a decision tree model.