🛡️ Churn Shield - Customer Churn Prediction App

🎬 See It in Action

Watch a 30-second demo of Churn Shield's 3 core pages:

(Click to play video)

📊 Overview

Churn Shield is an intelligent customer churn prediction application built with Streamlit and machine learning. It helps businesses identify customers who are likely to discontinue their services, enabling proactive retention strategies.

🎯 Key Features

• 🔮 Smart Predictions: Predict customer churn with high accuracy using Random Forest classifier
• 📈 Visual Insights: Comprehensive data visualization and customer behavior analysis
• 🎛️ Interactive Interface: User-friendly web interface for easy data input and results
• 📊 Real-time Analytics: Instant prediction results with confidence scores

🚀 Demo

The application provides three main sections:

1. 🏠 Home: Introduction and dataset information
2. 🔍 Insights: Visual analysis of customer behavior patterns
3. 🎯 Predict: Interactive prediction tool for individual customers

📊 Power BI Dashboard

🛠️ Technology Stack

• Frontend: Streamlit
• Machine Learning: scikit-learn (Random Forest Classifier)
• Data Processing: Pandas, NumPy
• Visualization: Matplotlib, Seaborn
• Model Persistence: Pickle

📋 Prerequisites

Before running the application, ensure you have Python 3.8+ installed on your system.

⚡ Quick Start

1. Clone the Repository

git clone https://github.com/mustafarezk12/churn-shield.git
cd churn-shield

2. Install Dependencies

pip install -r requirements.txt

3. Run the Application

streamlit run app.py

The application will open in your default web browser at http://localhost:8501

📦 Installation

Create a requirements.txt file with the following dependencies:

streamlit>=1.28.0
pandas>=1.5.0
numpy>=1.24.0
scikit-learn>=1.3.0
matplotlib>=3.6.0
seaborn>=0.12.0
streamlit-option-menu>=0.3.6

🎮 How to Use

Making Predictions

1. Navigate to the Predict tab

2. Enter customer details:

   • Age: Customer's age (0-120)
   • Gender: Male or Female
   • Tenure: Months with the company (0-100)
   • Usage Frequency: Monthly usage count
   • Support Calls: Monthly support interactions
   • Payment Delay: Days of payment delay
   • Subscription Type: Basic, Standard, or Premium
   • Contract Length: Monthly, Quarterly, or Annual
   • Total Spend: Customer's total expenditure
   • Last Interaction: Days since last contact

3. Click Predict to get instant results with confidence scores

Viewing Insights

The Insights section provides:

• Customer demographic distributions
• Subscription and contract analysis
• Support call patterns
• Interactive dashboard visualizations

📊 Dataset Features

The model uses 12 key features to predict customer churn:

Feature	Description
CustomerID	Unique customer identifier
Age	Customer age
Gender	Customer gender
Tenure	Months as customer
Usage Frequency	Monthly usage count
Support Calls	Monthly support contacts
Payment Delay	Payment delay in days
Subscription Type	Service tier
Contract Length	Contract duration
Total Spend	Total customer expenditure
Last Interaction	Days since last contact
Churn	Target variable (0/1)

🧠 Model Performance

• Algorithm: Random Forest Classifier
• Features: 10 input features (excluding CustomerID and Churn)
• Preprocessing: Standardization using mean and standard deviation
• Output: Binary classification with probability scores

🎨 User Interface

The application features:

• Responsive Design: Works on desktop and mobile devices
• Interactive Navigation: Easy-to-use horizontal menu
• Visual Feedback: Color-coded prediction results
• Professional Styling: Clean and modern interface

📁 Project Structure

churn-shield/
├──Dataset
|    ├── customer_churn_dataset-testing-master.csv
|    ├── customer_churn_dataset-training-master.csv
|
├──EDA&ML-MODel
     ├── Full_Customer_Churn(1).ipynb  # EDA & Trained/test ML model
├── Streamlit  # Main Streamlit application
|___ |___ app.py
|    |___ model(1).pkl # Trained ML model
|    |___ exported_data.csv  # Training dataset
|    |___ mean_std_values.pkl # Normalization parameters
|    |___ requirements.txt     # Python dependencies
├── images
|   ├── LOGO.jpg             # Application logo
|   ├── image.png            # Prediction page image
|   ├── dash.PNG             # Dashboard screenshot
└── README.md            # Project documentation

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

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

📝 License

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

👥 Authors

• Your Name - Initial work - YourGitHub

🙏 Acknowledgments

• Dataset providers for the customer churn data
• Streamlit team for the amazing framework
• scikit-learn contributors for the machine learning tools

📞 Support

If you have any questions or run into issues, please:

1. Check the Issues page
2. Create a new issue if your problem isn't already reported
3. Provide detailed information about your environment and the issue

---

Made with ❤️ and Python

⭐ Star this repository if you found it useful!