Customer Churn Classification on Telecommunication Customer Data Using the Voting Classifier Method

Abstract

Customer churn is a major challenge in the telecommunications industry due to its direct impact on revenue decline. This study aims to develop a customer churn prediction model using the Voting Classifier method, which combines Naive Bayes, Random Forest, and Logistic Regression algorithms. The dataset used is a public telecommunication customer dataset consisting of 7,043 entries with 20 attributes. The training process involved various train-test split ratios and class balancing using the SMOTE method. Model performance was evaluated using classification metrics such as accuracy, precision, recall, and F1-score. The results show that the Voting Classifier approach outperforms individual models, achieving an accuracy of 88.8%, precision of 93.61%, recall of 86.27%, and F1-score of 89.79%. This approach has the potential to support decision-making in customer retention strategies within the telecommunications sector.