Machine Learning helps businesses improve their cross-selling capabilities and increase customer loyalty. In fact, many companies nowadays deploy recommender systems, which infer their customers’ preferences to propose them personalized products or services. With such a system in place, an online groceries store can propose shopping baskets, a multimedia platform can recommend novel videos, or an investment platform can recommend custom portfolios to the customers.
But what if a regular shopper suddenly decided to try a vegan diet? What if a video consumer wants more sci-fi content? What if the investor wants to switch to long term and low-risk investments?
In this paper we discuss how to implement the recommender system that responds to the shifting preferences of its users. We demonstrate our implementation with a FinTech use case.