Porblem

there are many of loan applications provided to be scanned and give a response to those applicants whether they are qualified or not for taking a vehicle loan which takes lots of time to scan these huge amount of applications which lead to customers dissatisfaction and as a result losing many of our clients and wasting company's resources like (time, employees effort,...).

Project Goal

reducing the company's wasted resources and increasing customer satisfaction by reducing waiting time for each loan application.

Project strategy

Creating a logistic regressing machine learning model to predict whether the loan applicant will default or not based on some predictors, which will give the organization the advantage of initially filtering loan applications to give more focus to customers who have a high likelihood of getting a loan

About the dataset..

A dataset from L&T Financial Sevices contains information about vehicle loans from Indea. Each data row expresses a unique loan that was issued to a customer.

Model building workflow

1. Prepare Data

   • import Data

   • EDA

   • Feature engineering

   • Splitting Data
     

2. Build Model

   • Build Model

   • Model evaluation
     

3. Conclusion

   • Key takeaways and recommendations