🏦 Loan Approval Prediction System
An intelligent machine learning application that predicts loan approval status based on applicant information.
🎯 Model Performance
- Accuracy: 91.9%
- Precision: 93.0%
- Recall: 94.5%
- F1-Score: 93.7%
📊 Features Analyzed
The model evaluates 11 key factors:
- Credit Score (CIBIL) - Primary creditworthiness indicator
- Annual Income - Earning capacity
- Loan Amount - Requested funding
- Loan Term - Repayment duration
- Asset Portfolio:
- Residential property value
- Commercial property value
- Luxury assets value
- Bank savings/assets
- Education Level - Graduate vs Non-Graduate
- Employment Type - Self-employed vs Employed
- Number of Dependents - Financial obligations
🚀 How to Use
- Enter all applicant details in the form
- Click "Submit" to get instant prediction
- View the decision along with confidence score
- Try the example scenarios to understand model behavior
💡 Key Insights
The model reveals that loan approval is most strongly influenced by:
- CIBIL Score (750+ greatly increases approval chances)
- Debt-to-Income Ratio (loan amount vs annual income)
- Total Assets (demonstrates financial stability)
🛠️ Technology Stack
- Framework: Python, scikit-learn
- Algorithm: Logistic Regression
- Interface: Gradio
- Deployment: Hugging Face Spaces
- Dataset: 4,269 loan applications
📈 Dataset Information
- Total Samples: 4,269 applications
- Approved: 62.2% (2,656 cases)
- Rejected: 37.8% (1,613 cases)
- Features: 11 numerical and categorical variables
⚠️ Disclaimer
This model is for educational and demonstration purposes only. Real-world loan approval decisions involve:
- Additional documentation verification
- Legal compliance checks
- Human judgment and discretion
- Company-specific policies
- Regulatory requirements
This tool should NOT be used as the sole basis for actual financial decisions.
📝 About
This project demonstrates:
- End-to-end machine learning workflow
- Data preprocessing and feature engineering
- Model training and evaluation
- Production deployment
- User-friendly interface design
Created as a portfolio project to showcase ML and deployment skills.
🤝 Connect
- GitHub: [ abhijain2402 ]
- LinkedIn: [ www.linkedin.com/in/abhi-jain-901a42285 ]
Try it out and see how different factors affect loan approval! 🎓
