Open Conference Systems, 50th Scientific meeting of the Italian Statistical Society

Font Size: 
Customer Churn prediction based on eXtreme Gradient Boosting classifier
Matteo Borrotti

Last modified: 2018-04-27


Nowadays, Machine Learning (ML) is a hot topic in many different fields. Marketing is one of the best sectors in which ML is giving more advantages. In this field, customer retention models (churn models) aim to identify early churn signals and recognize customers with an increased likelihood to leave voluntarily. Churn problems fit in the classification framework, and several ML approaches have been tested. In this work, we apply an innovative classification approach, eXtreme Gradient Boosting (XGBoost). XGBoost demostrated to be a powerful techniques for churn modelling purpose applied to the retail sector.

Full Text: PDF