Samenvatting

Personalization of media services is gaining more and more traction, e.g., through the rise of personalization driven by recommender systems across media outlets. At the same time, we see a general rise in distrust and skepticism around the collection and processing of personal data, spurred by policy changes such as the introduc- tion of the GDPR, data breach incidents, and the rise of privacy concerns in general. We feel it is of central importance to be trans- parent about the information we collect, and the ways we use it. In this position paper we motivate the importance of enabling transparency through explaining our recommender system. More specifically, we aim to explain the inferred user profiles that are cen- tral to content-based recommender systems. We describe how user profile explanations can contribute to, or enable different aspects of our recommender system; transparency to help users better under- stand the inner workings of the recommender system, scrutability to allow users to provide explicit feedback on the internally con- structed user profiles, and self-actualization to support users in understanding and exploring their personal preferences. Finally, we believe that user profile explanations can find novel and interesting explanations as an end in itself.