🏦 Loan Default Prediction — Streamlit Web App

An interactive data analytics and machine-learning application built with Streamlit for exploring a loan-default dataset and predicting borrower risk in real time.

---

📋 Table of Contents

• Overview
• Features
• Dataset
• Project Structure
• Installation
• Usage
• Screenshots
• Technologies Used
• Contributing
• License

---

Overview

This project provides an end-to-end workflow for Loan Default Prediction:

1. Explore — Rich exploratory data analysis (EDA) with 10+ interactive chart types.
2. Model — Train and compare three ML algorithms (Random Forest, Gradient Boosting, Logistic Regression).
3. Predict — Enter applicant details and receive an instant default-risk prediction with a probability gauge.

The app is designed with a premium, modern UI featuring gradient themes, animated cards, and smooth interactions.

---

Features

🏠 Home Page

• Hero section with a gradient banner and animated feature cards.
• At-a-glance dataset snapshot (total records, features, default rate).

📊 Exploratory Data Analysis (EDA)

Visualization	Description
Data Overview	Raw data preview, column types, missing-value audit, descriptive statistics
Histograms + Box Plots	Distribution of any numerical feature with marginals
Violin Plots	Density + box plot overlay for numerical features
Small Multiples	All numerical distributions side-by-side
Bar & Donut Charts	Counts and proportions for categorical features
Stacked Default Bars	Default rate breakdown per category
Correlation Heatmap	Pairwise Pearson correlation matrix
Scatter Plot Studio	Interactive X/Y/Color selectors
Box Plots vs Target	Feature distributions split by default status
Pair Scatter Matrix	Pairwise scatter plots for top features
Grouped Mean Bars	Average feature values by default status

🤖 Machine Learning Modeling

Feature	Description
3 Algorithms	Random Forest · Gradient Boosting · Logistic Regression
Configurable Split	Adjustable test-size slider (10 %–40 %)
Evaluation Metrics	Accuracy · Precision · Recall · F1 Score
Confusion Matrix	Annotated heatmap
ROC Curve	Area under the curve visualization
Classification Report	Full precision / recall / F1 per class
Feature Importance	Top-15 bar chart (tree-based models)

🔮 Prediction Playground

• Dynamic input form with sliders and dropdowns.
• Real-time default probability with a visual gauge indicator.
• Clear risk verdict (✅ Low Risk / ⚠️ High Risk).

---

Dataset

The dataset (Loan_default.csv) contains 255,348 records and 18 columns:

Column	Type	Description
LoanID	String	Unique loan identifier (dropped during preprocessing)
Age	Integer	Borrower's age
Income	Integer	Annual income
LoanAmount	Integer	Requested loan amount
CreditScore	Integer	Credit score (300–850)
MonthsEmployed	Integer	Months at current job
NumCreditLines	Integer	Number of open credit lines
InterestRate	Float	Loan interest rate (%)
LoanTerm	Integer	Loan term in months (12, 24, 36, 48, 60)
DTIRatio	Float	Debt-to-income ratio
Education	Category	High School · Bachelor's · Master's · PhD
EmploymentType	Category	Full-time · Part-time · Self-employed · Unemployed
MaritalStatus	Category	Single · Married · Divorced
HasMortgage	Category	Yes / No
HasDependents	Category	Yes / No
LoanPurpose	Category	Home · Auto · Education · Business · Other
HasCoSigner	Category	Yes / No
Default	Binary	Target — 0 = No Default, 1 = Default

---

Project Structure

Loan_default.csv/
├── app.py              # Main Streamlit entry point & navigation
├── eda.py              # EDA visualizations (10+ chart types)
├── model.py            # ML training, evaluation & prediction
├── utils.py            # Data loading & preprocessing utilities
├── Loan_default.csv    # Source dataset (255k rows)
├── requirements.txt    # Python dependencies
└── README.md           # This file

---

Installation

Prerequisites

• Python 3.10+ installed on your system.
• (Optional) A virtual environment tool (venv, conda, etc.).

Steps

# 1. Clone or download the project
cd "path/to/Loan_default.csv"

# 2. (Recommended) Create a virtual environment
python -m venv venv
venv\Scripts\activate        # Windows
# source venv/bin/activate   # macOS / Linux

# 3. Install dependencies
pip install -r requirements.txt

---

Usage

# Launch the Streamlit app
streamlit run app.py

The app will open automatically at http://localhost:8501.

Use the sidebar to navigate between pages:

Page	What to do
🏠 Home	Read the overview and check the dataset snapshot
📊 Exploratory Analysis	Select features, switch tabs, explore visualizations
🤖 ML Modeling	Pick an algorithm → Train → Evaluate → Predict

---

Screenshots

Launch the app and explore each page to see the premium UI in action!

• Home — Gradient hero banner, feature cards, KPI metrics
• EDA — Interactive Plotly charts with tabs for Overview / Univariate / Bivariate
• ML Modeling — Model training, confusion matrix, ROC curve, feature importance, prediction gauge

---

Technologies Used

Technology	Purpose
Streamlit	Web application framework
Pandas	Data manipulation and analysis
NumPy	Numerical computing
Plotly	Interactive visualizations
Matplotlib	Static plotting
Seaborn	Statistical visualizations
scikit-learn	Machine learning pipeline

---

Contributing

Contributions are welcome! Feel free to:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Built with ❤️ using Streamlit • 2026