Using predictive models to approve merchandise lease applications in real time Snap Finance is a digital finance company that provides merchandise lease financing to brick and mortar as well as e-commerce merchants. The Snap lease-purchase agreement is an innovative financial product, which gives the 40% of consumers with poor credit an alternative to payday loans. The Challenge When consumers apply on Snap’s web site or in stores, Snap uses predictive models to decide whether to approve the lease. In order to make its predictions more accurate, Snap developed more sophisticated machine learning models in R, and stood up a process to continuously improve these models through rapid iteration. Snap needed a way to integrate its new, R-based models into its core web application, which is buil

Join for free to read