Towards incorporating ethics in recommendation systems.

Summary Most of product recommender systems are based on artificial intelligence algorithms using machine learning or deep learning techniques. One of the current challenges is to avoid negative effects of these product recommender systems on customers (or prospects), such as unfairness, biais, discrimination, opacity, encapsulated opinion in the implemented recommender systems algorithms. This paper is about the challenge of fairness. We define the concept and present some measures of fairness. Next, we will present a new predictive model, to which we plan to incorporate equity criteria. Using a dataset from the entertainment industry, we measure the fairness for each method and compare the results.