🎯 Customer Churn Prediction

Made with ❤️ by Sheick

📋 Overview

An advanced machine learning application that predicts customer churn probability using ensemble methods and provides personalized insights through an interactive Streamlit dashboard.

🌟 Features

• Multi-Model Prediction System

  • Random Forest
  • XGBoost
  • K-Nearest Neighbors
  • Ensemble Voting Classifier

• Interactive Dashboard

  • Real-time predictions
  • Dynamic visualization
  • Customer profile analysis
  • Personalized recommendations

• Advanced Analytics

  • Probability gauges
  • Feature importance analysis
  • Model comparison charts
  • AI-powered explanations

🛠️ Technical Stack

• Frontend: Streamlit
• Backend: Python 3.12
• ML Libraries:
  • scikit-learn
  • XGBoost
  • pandas
  • numpy
• API Integration: Groq API for AI explanations
• Data Visualization: Plotly

🚀 Installation

1. Clone the repository

git clone https://github.com/yourusername/Customer_Churn.git
cd Customer_Churn

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 environment variables

cp .env.example .env
# Add your GROQ_API_KEY to .env

5. Run the application

streamlit run main.py

📊 Model Performance

Model	Accuracy	F1-Score	ROC AUC
XGBoost	0.859	0.85	0.89
Random Forest	0.854	0.84	0.88
KNN	0.832	0.82	0.85

🔍 Project Structure

Customer_Churn/
├── main.py              # Main Streamlit application
├── utils.py             # Utility functions
├── models/             # Trained ML models
├── churn.csv           # Dataset
├── requirements.txt    # Project dependencies
└── README.md          # Project documentation