Nowadays, there is no shortage of options for customers when choosing where to put their money. As a result, customer churn and engagement have become one of the top issues. With the increase in the number of service providers for the same targeted population, there is a need for service providers to try to find the changing customer behaviour and their rising expectations to retain them. Various studies have proposed customer churn. Data mining was routinely used to predict telecom customer attrition. Most researchers have compared and proposed different approaches for the prediction of customer churn, though some of the Machine learning (ML) algorithms used were unable to provide the performance needed to identify customer churn. Therefore, this paper presents a comparative analysis of Support Vector Machine (SVM), Decision Tree (DT) and Random Forest (RF) in the Telecommunications Dataset. To prepare the dataset for machine learning algorithms, chi-square was used for feature selection to select the most informative features from the original dataset. We validate our model using a ten-fold cross-validation approach to test the performance of our models. RF model performed better than other models in terms of accuracy (94%), precision (94%) and F-measure (94%), respectively. Additionally, we compared our results with existing models that used the same dataset; the proposed strategy outperformed them. Keywords: Customer Churn, Telecommunication, Machine Learning, Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF)
Hambali et al. (Thu,) studied this question.