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 technique for churn modelling purpose applied to the retail sector.

Hassan Elbedawi Omar, M., Borrotti, M. (2018). Customer churn prediction based on eXtreme gradient boosting classifier. In Book of Short Papers SIS 2018 (pp.775-780).

Customer churn prediction based on eXtreme gradient boosting classifier

Borrotti, M
2018

Abstract

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 technique for churn modelling purpose applied to the retail sector.
paper
XGBoost; Gradient Boosting: Classification problem; Churn model
English
Scientific Meeting of the Italian Statistical Society
2018
Book of Short Papers SIS 2018
9788891910233
2018
775
780
none
Hassan Elbedawi Omar, M., Borrotti, M. (2018). Customer churn prediction based on eXtreme gradient boosting classifier. In Book of Short Papers SIS 2018 (pp.775-780).
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/214768
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact