Predicting-Loan-Default

In this data science project, we will predict borrowers chance of defaulting on loans by building a default prediction model.

Default (Banking Terminology): The borrower unable to repay the loan

Techniques Used

1. Decision Tree Classifier
2. Random Forest Classifier
3. Logistic Regression
4. Support Vector Machines
5. Artifical Neural Networks

Out of the five techniques used in this project, it is found that Artifical Neural Networks gives the highest precision of 82%.

Weights file for model1: two_layer_credit_score_model.h5

Dataset

Name: Credit_Scoring.csv

First five rows of the dataset

All in One Notebook: Default_Predection.ipynb

Credit Score Prediction

You can go ahead and try Credit Score Prediction using the "German Credit Dataset".

Credit Score (Banking Terminology): A credit score is a number between 300–850 that depicts a consumer's creditworthiness. The higher the score, the better a borrower looks to potential lenders. A credit score is based on credit history: number of open accounts, total levels of debt, and repayment history, and other factors.

It Is Similar to Loan Prediction.

The German Credit Dataset and its description is in the folder named "Credit Score"