Maurizio Ferrari Dacrema, PhD candidate in Information Technology, and Paolo Cremonesi, of the System Architectures research group of DEIB, have won the ACM Europe Council Best Paper Award at the 13th ACM Recommender Systems conference. The paper "Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches" presents a systematic analysis of deep learning recommender systems published in recent years at top level conferences. The paper identifies 18 relevant articles, only 7 have reproducible results and 6 of them can be outperformed by much simpler non-neural methods. Overall, the paper sheds light on a number of potential problems in today’s machine learning scholarship and calls for improved scientific practices in this area. The ACM Recommender Systems conference (RecSys) is the premier international forum for the presentation of new research results, systems and techniques in the broad field of recommender systems. Recommender systems are an information filtering tool whose aim is to help the user in exploring vast catalogues of information, products and news available in this age of information and booming e-commerce. The suggestions received by the user are be personalized based upon his/her previous interactions with the system, preferences and other personal characteristics.


ACM Europe Council: https://europe.acm.org/