This project is designed to help businesses gain valuable insights into customer behavior by predicting whether a customer is likely to leave within a specific period of time. Leveraging the power of XGBoost classifier and the scalability of the Google Cloud Platform (GCP), our model provides actionable predictions to enhance customer retention strategies and optimize business outcomes. Reduced customer churn directly contributes to improved business profitability, customer loyalty, and long-term success.
- Predictive Model: We utilize the powerful XGBoost classifier to create a predictive model that analyzes customer data and assigns a churn prediction label.
- Feature Engineering: Our model is trained on a range of customer data attributes, including demographics, usage patterns, engagement metrics, and more.
- Scalable Deployment: The project is deployed on the Google Cloud Platform, allowing for seamless scalability to handle large volumes of customer data.
customer-churn is a Python project, to run it locally you need to install Python version 3.11, and Python dependencies are listed in requirements.txt (application main requirements), requirements-test.txt (test requirements), and requirements-dev.txt (requirements for development of the project).
We use sample open-source data from Kaggle for this project's proof of concept. You can use the same dataset to reproduce the model and get your application up and running.
For more information, visit dataset on Kaggle.
I welcome contributions from the community! If you have suggestions, improvements, or additional features to add, please feel free to contact me via email (soltaniradali@gmail.com) or submit a pull request.