Customer Churn Prediction

Machine learning project to predict customer churn (Telco Customer Churn) using PostgreSQL/MongoDB as data sources, with exploratory analysis, preprocessing, model training, and dashboard.

🚀 Live Demo

Try the Live Application

📊 Project Overview

This project analyzes customer data to predict whether a customer is likely to churn (leave the service). It uses machine learning algorithms to identify patterns in customer behavior and provides actionable insights for customer retention strategies.

✨ Features

• Interactive Web Interface: User-friendly Streamlit application for real-time predictions
• Multiple Data Sources: Support for PostgreSQL and MongoDB databases
• Comprehensive Analysis: Exploratory data analysis with visualizations
• Machine Learning Models: Trained models for accurate churn prediction
• Data Processing Pipeline: Automated data preprocessing and feature engineering
• Model Explanations: AI-powered explanations for prediction results

🛠️ Technology Stack

• Frontend: Streamlit
• Backend: Python
• Databases: PostgreSQL, MongoDB
• Machine Learning: Scikit-learn, Pandas, NumPy
• Data Visualization: Matplotlib, Seaborn, Plotly
• Model Persistence: Pickle

📁 Project Structure

customer-churn-prediction/
├── app/
│   ├── customer_churn_app_explanation.py    # AI explanations for predictions
│   └── streamlit_app.py                     # Main Streamlit application
├── data/
│   ├── processed/
│   │   └── telco_processed.csv             # Cleaned and processed data
│   └── raw/
│       └── WA_Fn-UseC_-Telco-Customer-C... # Original dataset
├── models/
│   ├── best_churn_model.pkl                # Trained ML model
│   ├── database_loaders_guide.md           # Database setup guide
│   └── model.ipynb                         # Model development notebook
├── notebooks/
│   ├── EDA_postgresql_mongodb.ipynb        # Database analysis
│   └── EDA.ipynb                          # Exploratory data analysis
└── scripts/
    ├── loaders/
    │   ├── mongodb_data_loader.py          # MongoDB data operations
    │   ├── postgresql_data_loader.py       # PostgreSQL data operations
    │   └── raw_data_loader.py             # Raw data processing
    └── preprocessor/
        ├── data_preprocessor_csv.py        # CSV data preprocessing
        ├── mongodb_data_preprocessor.py    # MongoDB preprocessing
        └── postgresql_data_preprocessing.py # PostgreSQL preprocessing

🚀 Getting Started

Prerequisites

• Python 3.8+
• pip or conda package manager
• PostgreSQL (optional, for database features)
• MongoDB (optional, for database features)

Installation

1. Clone the repository

   git clone https://github.com/AbhijeetDasBakshi/customer-churn-prediction.git
   cd customer-churn-prediction

2. Create a virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Set up databases (optional)

   • Follow the guide in models/database_loaders_guide.md for PostgreSQL and MongoDB setup

Running the Application

1. Start the Streamlit app

   streamlit run app/streamlit_app.py

2. Access the application

   • Open your browser and navigate to http://localhost:8501

📈 Usage

Making Predictions

1. Input Customer Data: Enter customer information in the web interface
2. Get Prediction: Click predict to see churn probability
3. View Explanation: Review AI-generated explanations for the prediction
4. Analyze Results: Use insights for customer retention strategies

Data Analysis

• Exploratory Analysis: Run notebooks/EDA.ipynb for comprehensive data insights
• Database Analysis: Use notebooks/EDA_postgresql_mongodb.ipynb for database-specific analysis

🤖 Model Information

The project uses a trained machine learning model (best_churn_model.pkl) that analyzes various customer features:

• Customer Demographics: Age, gender, partner status
• Service Information: Contract type, payment method, tenure
• Usage Patterns: Monthly charges, total charges, services used
• Account Details: Paperless billing, automatic payments

---

---

🔧 Development

Adding New Features

1. Data Processing: Modify scripts in scripts/preprocessor/
2. Model Training: Update models/model.ipynb
3. Web Interface: Enhance app/streamlit_app.py

Database Integration

The project supports multiple data sources:

• CSV Files: Direct file processing
• PostgreSQL: Relational database integration
• MongoDB: NoSQL database support

📊 Data Sources

The project uses telecommunications customer data from kaggle with features including:

• Customer account information
• Demographic details
• Service subscriptions
• Billing and payment data
• Usage statistics

Data Source

🤝 Contributing

1. Fork the repository
2. Create a 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.

🤝 Acknowledgments

• Dataset provided by Kaggle
• Streamlit for the amazing web framework
• Scikit-learn for machine learning capabilities

📞 Support

If you have any questions or issues, please:

1. Create a new issue if needed. Check the Issues page
2. Contact: abhijeet.bakshi321@gmail.com

---

⭐ If you found this project helpful, please give it a star!