An open API service indexing awesome lists of open source software.

https://github.com/pythonbyte/default-loans-analysis

This repository displays a CRISP-DM approach to understand about how Loans enter on default status.
https://github.com/pythonbyte/default-loans-analysis

Last synced: 12 months ago
JSON representation

This repository displays a CRISP-DM approach to understand about how Loans enter on default status.

Awesome Lists containing this project

README

          

### Table of Contents

1. [Installation](#installation)
2. [Data](#data)
3. [Project Motivation](#motivation)
4. [File Descriptions](#files)
5. [Results](#results)
6. [Licensing, Authors, and Acknowledgements](#licensing)

## Installation

For this project, some main data science libraries were used and will be needed for properly running the notebook.

These are the libraries:

* numpy
* pandas
* matplotlib
* seaborn
* scikit-learn

## Data

The data used for this project can be found here on this publicly available repository of [Kaggle](https://www.kaggle.com/gauravduttakiit/loan-defaulter).

## Project Motivation

The area of Loans has a really interesting machine learning problem which is predicting who has a defaulted loan based on the customer's characteristics. This motivated me to pursue the answer to the following questions.

* Which type of contract is the most defaulted?
* How the default behavior is divided across the customer's social context.
* What are the factors that most relate a customer to default a loan?

## File Descriptions

The file distribution on this project is pretty straightforward, the CRISP-DM approach used to analyze the Loan dataset is on the default-loans-eda.ipynb notebook, and the rest of the files are license, readme, gitignore.

## Results

The results from this data analysis can be found on this vailable [Post](https://medium.com/@eduardommelgaco/this-data-analysis-will-make-you-rethink-how-loans-are-given-bee93bb8fb87).

## Licensing, Authors, Acknowledgements

If you want to chat about this analysis or other approaches you can find me on [Twitter](https://twitter.com/python_byte). Feel free to use this code for any reference you need. And a great thanks to Kaggle for having this community that provides great datasets and Udacity for enabling this project to happen.