The Investigation of the Advancement for Machine Learning-Based Telecommunication Customer Churn Prediction

Yi Zhang

2024

Abstract

It is important for companies to predict customer churn, as it helps reduce losses, and machine learning is crucial to this process, making its study extremely important. Ensemble Methods, Logistic Models, Classification Trees, and Nearest Neighbor Algorithms are all traditional machine learning methods, but they have inherent drawbacks. This encompasses overfitting, a situation where models achieve high accuracy on training data but struggle to perform effectively on unseen data; model complexity, making complex models difficult to interpret and maintain; and inefficiencies in handling large-scale data, often leading to poor performance with vast amounts of data. Conversely, advancements in deep learning highlight its strengths and areas where it surpasses traditional methods. This study offers an extensive perspective on telco churn prediction and artificial intelligence, delving into possible concerns like data protection, which aims to protect user data from exploitation. To address these challenges, several solutions are proposed, including collaborating with experts to ensure the accuracy and reliability of AI models, implementing stronger security measures to protect sensitive information, and utilizing techniques to mitigate data scarcity issues. Overall, this work offers an excellent review of the telco churn prediction field, highlighting key advancements and proposing solutions to existing challenges.

Download


Paper Citation


in Harvard Style

Zhang Y. (2024). The Investigation of the Advancement for Machine Learning-Based Telecommunication Customer Churn Prediction. In Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI; ISBN 978-989-758-726-9, SciTePress, pages 56-60. DOI: 10.5220/0013206400004568


in Bibtex Style

@conference{ecai24,
author={Yi Zhang},
title={The Investigation of the Advancement for Machine Learning-Based Telecommunication Customer Churn Prediction},
booktitle={Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI},
year={2024},
pages={56-60},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013206400004568},
isbn={978-989-758-726-9},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI
TI - The Investigation of the Advancement for Machine Learning-Based Telecommunication Customer Churn Prediction
SN - 978-989-758-726-9
AU - Zhang Y.
PY - 2024
SP - 56
EP - 60
DO - 10.5220/0013206400004568
PB - SciTePress