Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 14

A comparative examination of Machine Learning algorithms for credit risk analysis

(2020CSEAI36)

by
UJJWAL SINGH(2002901520122)
SHASHANK PATEL(2002901520102)
VISHAL SHARMA(2002901520124)
Under the Supervision of
Mrs. KHUSHBOO SAXENA
Asstt. Professor
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING - AI
ABESIT INSTITUTE OF TECHNOLOGY GHAZIABAD
Affiliated to
DR. APJ ABDUL KALAM TECHNICAL UNIVERSITY
(Formerly Uttar Pradesh Technical University, Lucknow)
INTRODUCTION
 In today's financial landscape, where data-driven decisions are
paramount, artificial intelligence is revolutionizing how we
evaluate creditworthiness.“ Imagine having a predictive model that
can process vast amounts of financial data swiftly and accurately.
 Your credit score is more than just a number; it's a financial
fingerprint that can significantly impact your financial life. It
influences your ability to secure loans, rent an apartment, obtain a
credit card, and even get a job. It's a fundamental measure of your
creditworthiness.
Methodlogy
Implementation

 The CSV data is from: https://www.lendingclub.com/info/prospectus.action


 Importing all the important Pacakages
• Create the List of All Columns Contain By CSV and create the
target to [Loan_Status] Column.
• Load the CSV file and perform Data cleaning

• Filter the Current status to Low_risk and others to High Risk


• Split Data into Training and Testing
• Resample the Training data with random forest and fit the model

• Calculate the Accuracy of the model


• The Imbalanced Classification report of Random Forest model

Now train the data with Adaboost model


• Calculating the balanced accuracy score

• The Imbalanced Classification report of Adaboost Model


• Accuracy Comparision of Random
Forest and Adaboost

 Random Forest model accuracy= 0.8731182795698925

 Adaboost model accuracy= 0.9424004649811102

 For this Data… Adaboost model shows more accuracy


then Random Forest.
REQUIREMENTS
HARDWARE
 RAM /Memory required: 4gb
 Hard disk capacity: 200mb
 Good Internet Connection (Up to 1Mbps)

SOFTWARE
 Jupyter Notebook
 Windows 10

Language used:
 Python
CONCLUSION

In closing, we've journeyed through the intricate world of credit score


analysis, shedding light on its importance, composition, and impact on our
financial lives. We've demystified the numerical credit scores and turned them
into understandable tools for financial empowerment.
THANK YOU

You might also like