Loan Eligibility Prediction Project 📊💸

Introduction

Loans constitute the fundamental business of banks, with the primary source of profit being the interest accrued on loans. Before granting a loan, financial institutions engage in a thorough process of verification and validation. Despite this, there remains uncertainty regarding whether the applicant possesses the capability to repay the loan without encountering difficulties. In this project, I construct a predictive model aimed at determining whether an applicant is likely to repay the loan to the lending company. 🏦💰

Problem Statement

Dream Housing Finance company specializes in providing all types of loans across urban, semi-urban, and rural areas. Upon receiving a loan application, the company undertakes a validation process to determine the customer's eligibility. This validation process is based on various customer details provided during the online application, including age, house ownership, working experience, intent of loan, income, loan amount, credit history, among others. To streamline and automate this loan eligibility process in real-time, the company aims to identify specific customer segments that qualify for loan amounts. By targeting these eligible customer segments, the company can optimize its loan offerings and better serve its clientele. 🏡💻

Top Insights

1. Age Distribution: Majority of borrowers are aged 20-35, indicating youthful loan applicants. 📊👨‍🎓

2. Housing Status: Half of our dataset resides in rented accommodations, while 40% have mortgaged homes. 🏠🔒

3. Loan Preference: Educational loans are most popular, closely followed by medical loans. Personal and venture loans show similar demand. 📚💉

4. Default Rates: While most borrowers have clean records, a small segment defaults 3-5 times. ⚠️

5. Loan Grades: Grades A and B dominate, reflecting lower risk. Grades C-G signify higher risk, with fewer instances. 📊🔢

6. Loan Amounts: Loans of $5000-$10000 are most common, followed by $300-$5000 loans. A significant group borrows over $15000. 💸💼

GitHub Repository

The complete project including Exploratory Data Analysis (EDA), model building, and a Streamlit application.

🙏 Thank You!

Thank you for exploring this project! If you have any questions, suggestions, or feedback, feel free to reach out.

Happy querying!