Churn analysis using deep convolutional neural networks and autoencoders
This addresses churn prediction for businesses by applying image-based deep learning to customer data, but it is incremental as it adapts existing methods to a new domain.
The paper tackled customer churn prediction by representing temporal behavioral data as images and using deep convolutional neural networks, achieving an AUC of 0.743 on a test dataset of over 6 million customers with up to 12 features. It also used autoencoders to analyze churn reasons, identifying actionable insights for preventing churn among specific user groups.
Customer temporal behavioral data was represented as images in order to perform churn prediction by leveraging deep learning architectures prominent in image classification. Supervised learning was performed on labeled data of over 6 million customers using deep convolutional neural networks, which achieved an AUC of 0.743 on the test dataset using no more than 12 temporal features for each customer. Unsupervised learning was conducted using autoencoders to better understand the reasons for customer churn. Images that maximally activate the hidden units of an autoencoder trained with churned customers reveal ample opportunities for action to be taken to prevent churn among strong data, no voice users.