An E-Commerce Customer Churn Study with Machine Learning

J. David Sukeerthi Kumar, A. Sindhu, J. Pranavi, D. Vasanthi, N. Harnitha

2025

Abstract

In the fast-evolving e-commerce environment, customer attrition is now a critical problem for the firm to maintain long term sustainability and profitability. The present study assumes an opensource e-commerce database and it tries to develop a robust predictive model of customer attrition. A study was conducted to determine the performance of various machine learning algorithms, including Random Forest, XG Boost, Light GBM, and Logistic Regression. SMOTE has been used for balancing the class imbalance and SHAP and LIME was used for making the models more interpretable. The Random Forest model had a very high predictability with a remarkable ROC AUC of 0.9850. Such findings can be seen as some of the churn predictors and proved out to be useful in how e-commerce business can deliberately use such data to reduce their customer turnover and implement the retaining policies.

Paper Citation

in Harvard Style

Kumar J., Sindhu A., Pranavi J., Vasanthi D. and Harnitha N. (2025). An E-Commerce Customer Churn Study with Machine Learning. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 875-879. DOI: 10.5220/0013874900004919

in Bibtex Style

@conference{icrdicct`2525,
author={J. Kumar and A. Sindhu and J. Pranavi and D. Vasanthi and N. Harnitha},
title={An E-Commerce Customer Churn Study with Machine Learning},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25},
year={2025},
pages={875-879},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013874900004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25
TI - An E-Commerce Customer Churn Study with Machine Learning
SN - 978-989-758-777-1
AU - Kumar J.
AU - Sindhu A.
AU - Pranavi J.
AU - Vasanthi D.
AU - Harnitha N.
PY - 2025
SP - 875
EP - 879
DO - 10.5220/0013874900004919
PB - SciTePress