Scribd
Upload
38 views7 pages

Final Project

The document discusses predicting vehicle loan defaults through machine learning. It describes the business problem of loan defaults causing losses and how analyzing key determinants can help minimize defaults. The objective is accurately predicting the probability of a borrower defaulting on their first loan payment. The dataset contains loan and borrower details and the target is a binary variable of loan default. Exploratory data analysis shows most are non-defaulters and higher loan-to-value ratios increase default probability.

Uploaded by

Anjana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
38 views7 pages

Final Project

The document discusses predicting vehicle loan defaults through machine learning. It describes the business problem of loan defaults causing losses and how analyzing key determinants can help minimize defaults. The objective is accurately predicting the probability of a borrower defaulting on their first loan payment. The dataset contains loan and borrower details and the target is a binary variable of loan default. Exploratory data analysis shows most are non-defaulters and higher loan-to-value ratios increase default probability.

Uploaded by

Anjana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

MACHINE LEARNING

PROJECT
BY S ANJANA
INTRODUCTION

• Individuals worldwide depend on banks to


lend them loans for various reasons to help
them overcome their financial constraints
and achieve some personal goals.

• Loan is the major source of income for the


banking sector and the biggest source of
financial risk for banks.
BUSINESS PROBLEM
• Financial institutions face substantial losses when borrowers default on vehicle loans,
prompting them to become more cautious in their lending practices and resulting in higher
rates of rejection for loan applications. This situation highlights the importance of
understanding the factors that contribute to vehicle loan defaults. In response, there is a
growing demand for an improved credit risk scoring model to better assess the likelihood
of default and make more informed lending decisions. This study aims to identify the key
determinants of vehicle loan defaults, providing valuable insights for financial institutions
seeking to enhance their risk assessment processes.

• Analysing this will help the financial institutions to minimize the loan default rates,
which helps us gain a customer base by providing the loans to the correct persons.
PROJECT OVERVIEW
• The objective of the project is to accurately predict the probability of
loanee/borrower defaulting on a vehicle loan in the first EMI (Equated Monthly
Instalments) on the due date.

• The dataset contains the following data: Loan Details Such as the loan amount, loan/value,
date of disbursal, and other related details. Information of the Loanee such as the ID,
date of birth, location, number of accounts, and other related details. By predicting the
candidates who might get involved with a loan default, financial institutions can take steps
to reduce the default by canceling the loan or by any other means.
About the Data
• Target Variable -
The response variable is loan_default which is given as:
The target variable has two classes and hence it is a binary classification problem. The
goal is to predict whether or not it will payment default on the first EMI or due date by
the customer.
EDA

• The count of non-defaulters is more than


defaulters where 0 (78.29%) is a non-defaulter
and 1 (21.71%) is a defaulter in the Loan Default
count plot.
•From the distribution plot, we see that as the
LTV increases the persons probability of
defaulting the loan also increases. Therefore,
higher the LTV higher are the chances of a
person defaulting the loan.
THE END

You might also like