Prediction and Analysis of Customer Churn at Telkomsel Using Machine Learning Approach

Abstract

Customer churn is one of the main problems in the telecommunications industry, including Telkomsel, the largest cellular operator in Indonesia. This study aims to build a classification model to predict customer churn and analyze the factors influencing churn using the CRISP-DM approach. Data was obtained through an online questionnaire from 100 respondents who are active students of Universitas Negeri Surabaya. The research process includes stages of data preparation (normalization, encoding, and removal of irrelevant attributes) and the application of classification algorithms such as Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors, Support Vector Machine, and Naïve Bayes. Evaluation was carried out using metrics such as accuracy, precision, recall, and F1-Score. The results show that Random Forest is the best algorithm with an F1-Score of 87.50% on an 80:20 data ratio. Feature analysis indicates that the attribute of previous churn status has the greatest influence on churn prediction