Building a Cloud-based Regression Model to Predict Click-through Rate in Business Messaging Campaigns

The goal of the research presented here is to describe an innovative approach to predicting the impact of a business messaging campaign, by estimating the percentage of message recipients who will engage with a message. The motivation is to facilitate business marketers to address the problem of estimating the return on investment coming from a potential messaging campaign. The presented solution relies on the processing of large scale business data, taking into account state-of-the-art predictive algorithms, GDPR compliance requirements, and the challenge of increased data security and availability. In this paper we discuss the design of the core functional components of a system that could make this possible, which encompasses predictive analytics, data mining and machine learning technologies in a cloud computing environment.