<![CDATA[The rapid growth of the telecommunications industry has increased competition among companies for customers. As a result, customers often switch to other services or terminate their subscriptions. Retaining customers is very important as it is 10 times cheaper than acquiring new customers. This study compares Random Forest (RF) and Convolutional Neural Network (CNN) algorithms in predicting customer switching, using Correlation-based Feature Selection (CFS) and Recursive Feature Elimination (RFE) for data partitioning. Model evaluation using Confusion Matrix and Area Under Curve (AUC). The evaluation results show that the performance of CNN models with optimization parameters is superior. Using the CFS dataset, the test data evaluation results yielded an accuracy of 98%, AUC of 0.96, precision of 99%, recall of 92%, and F1-score of 96%. The best tuning result for CNN is achieved with three combinations of filter and kernel sizes {[64, 7], [32, 3], [16, 2]} and a pool size of 2. A limitation of this research is determining how to compare the two algorithms being evaluated effectively. Both use different approaches, namely Supervised Learning and Deep Learning.]]>
Copyrights © 2025
