International Journal of Engineering and Information Systems (IJEAIS)   Year: 2023 | Volume: 7 | Issue: 4 | Page No.: 42-46 Predictive Modeling Of Customer Churn In Telecom Industry Using Machine Learning Algorithms Download PDF Muhammad Arfan, M Somantri, E Handoyo Abstract: Customer churn in the telecommunications industry. A large dataset of customer information was collected and preprocessed to prepare it for analysis. The dataset was split into training and testing datasets, and several machine learning algorithms were compared to identify the best-performing algorithm for predicting customer churn. The results showed that a neural network algorithm achieved the highest accuracy of 85%. Further analysis revealed that customer service interactions and the length of time since the customer's last purchase were the most important factors contributing to churn. Customers who had negative interactions with customer service and those who had not made a purchase in a long time were more likely to churn. These findings suggest that improving customer service interactions and targeted marketing campaigns to re-engage inactive customers can have a significant impact on customer retention. Overall, the study demonstrates the potential of predictive modeling using machine learning algorithms to reduce customer churn and improve customer retention. The findings have important implications for businesses in the telecommunications industry and beyond, highlighting the importance of analyzing customer data and taking proactive steps to retain customers. In today's competitive business landscape, customer retention is critical for the success of any company. Predictive modeling of customer churn using machine learning algorithms can help businesses identify customers who are at risk of churning and take appropriate measures to retain them. In this study, we explore the effectiveness of various machine learning algorithms in predicting customer churn using data collected between 2018 and 2021. We follow the CRISP-DM methodology and evaluate the performance of several algorithms, including logistic regression, decision trees, random forests, and neural networks. Our results demonstrate the effectiveness of machine learning algorithms in predicting customer churn, with an accuracy of 85% achieved using the logistic regression algorithm.