Recommender Systems: Beyond Machine Learning

Recommender systems help users find items of interest and help websites and marketers select items to promote. Today's recommender systems incorporate sophisticated technology to model user preferences, model item properties, and leverage the experiences of a large community of users in the service of better recommendations. Yet all too often better recommendations—at least by traditional measures of accuracy and precision—fail to meet the goal of improving user experience. This talk takes a look at successes and failures in moving beyond basic machine learning approaches to recommender systems to emphasize factors tied to user behavior and experience. Along the way, we explore approaches to combining human-centered evaluation with data mining and machine learning techniques.

Joseph Konstan

Joseph is Distinguished McKnight University Professor and Distinguished University Teaching Professor of Computer Science and Engineering at the University of Minnesota. He has been working in the field of recommender systems since 1995. He's published more than fifty research articles on the topic, holds five patents related to recommender systems, and co-authored the book Word of Mouse: The Marketing Power of Collaborative Filtering, one of the first books on the application of recommender systems to commercial systems. Konstan chaired the first ACM Conference on Recommender Systems, and has been active in ACM, including serving as President of ACM SIGCHI from 2003-2006; he is now starting his third term on the ACM Council. He co-founded Net Perceptions, Inc. in 1996. The company commercialized recommendation engines and had a variety of online and bricks-and-mortar companies among its customers, including Amazon.com. He is a Fellow of the ACM, has been elected to the SIGCHI Academy, and was part of the team that won the 2010 ACM Software Systems Award for the GroupLens Collaborative Filtering Recommender Systems.