A Comparison of Customer Churn Vector Embedding Models with Deep Learning

Abstract

In the telecommunication industry, deep learning has been utilized for churn prediction. Some companies have used sophisticated deep learning techniques to predict churn, which yielded satisfactory results. However, future studies are still required to evaluate several deep learning mechanisms as only SoftMax Loss has been used so far. By comparing customer churn vector embedding models with several methods, including SoftMax Loss, Large Margin Cosine Loss, Semi-Supervised Learning, and a combination of Large Margin Cosine Loss and Semi-Supervised Learning. The use of Large Margin Cosine Loss has been proven in face recognition which can increase the discrimination between vectors embedding in different classes. Understanding how mixing unlabeled and labeled input might alter developing algorithms and learning behavior that benefit from this combination are the goals of semi-supervised learning. This approach encouraged feature discrimination in customer behavior and improved the model's overall accuracy. Large Margin Cosine Loss in this study achieved 83.74% of the F1 Score compared to other methods. It was further demonstrated that the produced vectors for churn prediction are discriminative by examining the cluster's similarity and the t-SNE plot. The t-SNE visualization showed that the proposed model produces highly discriminative vectors with the Large Margin Cosine Loss model embedded vector being thicker than SoftMax Loss, Semi-Supervised Learning, and a combination of Large Margin Cosine Loss and Semi-Supervised Learning model churn clusters.