🧠 Customer Churn Prediction Platform

A full-stack, user-centric machine learning platform to predict customer churn with deep insights, model customization, and educational resources.

Built using React, Flask, and Scikit-learn, this app enables real-time and batch predictions, lets users train custom models, and explains predictions using SHAP and business logic visualizations.

---

🚀 Features

1. 🔍 Single Customer Prediction

• Fill out a form with customer details
• Get churn prediction in real time
• Visual insights using SHAP values and business logic explanations (via Recharts)

2. 📂 Batch Prediction

• Upload a CSV of multiple customers
• Download CSV with churn predictions and probability scores

3. 🧠 Model Manager

• Upload your own dataset to train a custom model
• Compare performance with the base model (Random Forest)
• Choose the preferred model for future predictions

4. 🏠 Home (Knowledge Hub)

• Short guides, tooltips, and explanations of churn concepts
• FAQs and actionable strategies based on predictions
• Use case examples and optional guided tour

---

🛠️ Tech Stack

• Frontend: React + Recharts (for visualizations)
• Backend: Flask (REST API)
• ML Pipeline:
  • Random Forest Classifier
  • SHAP for explainability
  • Business logic rule engine
  • Data balancing (SMOTEENN)
  • Data cleaning & preprocessing
• Other Tools: Pandas, Scikit-learn, Imbalanced-learn

---

📸 Preview

Live demo Link- https://customer-churn-analyzer-r19l.onrender.com

• Single prediction with insights
  

  

  

• Batch prediction result

  

  

  

• Model training & comparison
  

  

  

  

  

• Knowledge Hub interface
  

  

  

  

  

  

---

📊 ML Model Info

• Default Model: Trained on original dataset with Random Forest
• Custom Model Support: Upload CSV to retrain
• Balancing: SMOTEENN for class imbalance
• Explainability: SHAP + Business logic
• Comparison Metrics: Accuracy, Precision, Recall, F1-Score

---

📚 Knowledge Hub Includes

• What is Churn?
• How does our model work?
• How to interpret predictions?
• Business actions to reduce churn
• FAQs and example use cases

---

🧪 How to Run Locally

Frontend (React)

cd frontend
npm i
npm run dev

Backend (Flask)

cd backend
python -m venv venv
source venv/bin/activate  # or venv\Scripts\activate on Windows
pip install -r requirements.txt
python app.py

📄 License

MIT License — feel free to use, fork, and contribute!

---

🤝 Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you'd like to change.

---

🙌 Acknowledgements

Thanks to the open-source community for tools like SHAP, SMOTEENN, and Recharts and Kaggle for providing dataset.

---