Evaluating Recommender Systems, seminar by Alan Said, CWI Amsterdam

The talk will be broadcast through Bambuser.


In this talk I will present some of the work on recommender systems evaluation I have been, and am currently, working on. I will discuss contextual similarities and differences between information retrieval and recommender systems as well as some of the fallbacks in today’s state-of-the-art evaluation techniques. The talk will focus on recommendation contexts and what evaluation models are applicable in which scenarios and settings.


Alan Said is a postdoctoral researcher and Marie Curie fellow at Centrum Wiskunde & Informatica in Amsterdam, The Netherlands. He is working on Information Retrieval, Information Access Systems, and Recommender Systems specifically focusing on evaluation. Prior to joining CWI, he received his PhD from the Technische Universität Berlin on the topic of the accuracy and utility of recommender systems. He has published in venues such as ACM RecSys, CSCW, UMAP, IUI, TIST, etc. and has been involved in the organization of several workshops, journal special issues, hackathons and conferences related to recommender systems and personalization.

Contact: Lucian Leahu, Mobile Life Center/KTH, lucian at sics.se

Wednesday, February 12, 2014, 10:00 to 11:00
SICS, room Knuth
Isafjordsgatan 22