Algorithmic development in recommender systems goes hand in hand with evaluation. For an initial period of time the recommendation problem was addressed as an issue of matching user preferences as accurately as possible. It was soon understood that while certainly a basic requirement, accuracy alone is not enough to procure useful recommendations. Novelty and diversity, for instance, have been identified as key properties to draw user value out of accurately predicted user interests. We will review notions, motivations, different angles and metrics developed in such broader perspectives, as well as some algorithmic implications. Time permitting, we will relate this outlook to recent advances in understanding and dealing with biases in recommender system evaluation and algorithms.
Short Bio
Pablo Castells is an Associate Professor at the Computer Science Department of the Universidad Autónoma de Madrid, Spain, where he leads the Information Retrieval group. His research interests are in the fields of recommender systems and information retrieval, dealing with models, theory, algorithms and evaluation. His most recent research focus in these areas includes novelty and diversity, algorithmic and evaluation bias, and interactive recommendation. His research contributions can be found in the main conferences and journals in the field, where he also co-organized workshops and tutorials on the aforementioned topics (e.g. at ACM RecSys, ACM SIGIR, ACM WSDM, and WWW). Pablo served as PC co-chair of RecSys 2016, and has been appointed general co-chair of ECIR 2020 and PC co-chair of SIGIR 2021.