Learning latent representations of bank customers with the Variational Autoencoder
作者：
Highlights：
• It is possible to steer the latent space of the Variational Autoencoder.
• Latent representations that learn the customers’ creditworthiness.
• Well-defined clustering structures with statistically different default probabilities.
• Our method enables visualization and suggests the number of clusters.
• The proposed methodology generalizes to unseen customers.
摘要
•It is possible to steer the latent space of the Variational Autoencoder.•Latent representations that learn the customers’ creditworthiness.•Well-defined clustering structures with statistically different default probabilities.•Our method enables visualization and suggests the number of clusters.•The proposed methodology generalizes to unseen customers.
论文关键词：Variational Autoencoder,Data representations,Clustering,Machine learning
论文评审过程：Received 10 April 2019, Revised 20 August 2020, Accepted 14 September 2020, Available online 15 September 2020, Version of Record 18 September 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114020
