E-Commerce Customer Churn Prediction

---

📌 Project Overview

This project aims to predict whether customers will churn or not by utilizing machine learning techniques. The model is built on the XGBoost algorithm, as it provided the best performance among tested models, achieving an accuracy score of 97%.

A simple and interactive Web application has been developed to provide a user-friendly interface for predictions. Link: https://ecommerce-churn-prediction-ml.streamlit.app/

---

🚀 Features

• Data preprocessing and analysis using Python.
• Visualization of dataset trends and insights.
• Machine Learning model for Customer Churn detection.
• XGBoost algorithm chosen for best accuracy.
• Interactive UI for real-time prediction.

---

🛠️ Technologies & Libraries Used

• Python
• Google Collab
• Streamlit (for UI)
• Scikit-learn (for ML model)
• Pandas (data handling)
• NumPy (numerical operations)
• Seaborn & Matplotlib (data visualization)

---

📊 Model Performance

• Algorithm Used: XGBoost
• Accuracy Achieved: 97%

---

📂 Project Structure

ECommerce_Customer_Churn_Prediction-ML_Model/

• E_Commerce_Customer_Churn_Prediction.ipynb # Google colab with model training
• app.py # ui of model
• requirements.txt # Dependencies
• README.md # Project documentation
• dataset.csv # Dataset used for training

---

⚡ How to Run the Project

1. Install dependencies:

   pip install -r requirements.txt

2. Run the Streamlit app:

   streamlit run app.py

---

📌 Future Improvements

• Experiment with other algorithms (Logistic Regression, Random Forest, etc.).
• Improve dataset quality and size for better accuracy.
• Deploy the model on cloud platforms like Heroku or AWS.

---

✨ Author

Developed by Shalini 💻

---