همایش بین المللی هوش مصنوعی

صفحه اصلی / اولین همایش بین المللی هوش مصنوعی

A Comprehensive Approach to Predicting Customer Churn with XGBoost

نویسندگان :

Reza Najari¹ Mehdi Sadeghzadeh²

1- M.Sc. Student, Department of Computer Engineering, Science and Research Branch, Islamic Azad University Tehran, Iran 2- Associate Professor, Department of Computer Engineering, Science and Research Branch, Islamic Azad University Tehran, Iran

کلمات کلیدی :

customer churn،XGBoost،machine learning،predictive modeling

چکیده :

Abstract--Customer churn is a significant challenge for businesses, leading to substantial financial losses when dissatisfied customers switch to competitors. Machine learning (ML) and deep learning (DL) methods have been increasingly employed to address this issue; however, achieving high accuracy and minimizing false predictions remain critical challenges. This study leverages the XGBoost model, a robust algorithm, for predicting customer churn. The model is optimized through parameter tuning and the SMOTE technique to address data imbalance. Experimental results demonstrate that the proposed approach achieves an overall accuracy of 85.89%, outperforming several existing methods. Additionally, this research compares the proposed model with a baseline study utilizing a BiLSTM-CNN hybrid approach. The findings highlight that a well-optimized XGBoost model offers superior predictive performance and serves as a valuable tool for businesses in managing and mitigating customer churn effectively.