RecSys 2016 Preview
I’m looking forward to going back to RecSys this year, reconnecting with old colleagues and meeting some new ones.
I also am involved with a couple of papers this year and will be presenting one, as well as serving as publicity co-chair. On Sunday morning, I will be presenting ‘Behaviorism is Not Enough: Better Recommendations through Listening to Users’, a paper I wrote with Martijn Willemsen for the Past, Present, and Future of Recommender Systems track. The central thesis of our paper is that listening to users’ explicitly articulated goals and preferences is an important aspect to developing compelling recommender experiences, particularly as recommenders move towards helping to solve important problems in the future. We offer both philosophical arguments for this idea, grounded in participatory design and respect for users, and practical examples of kinds of recommendation experiences that cannot be effectively realized by only examining users’ behavior. Jennifer will be presenting our paper ‘First Do No Harm: Considering and Minimizing Harm in Recommender Systems Designed for Engendering Health’ at the Workshop on Engendering Health with RecSys on Thursday. This paper lays out various ways in which well-meaning health recommendations could bring harm to their users, and argues that health recommender applications need to carefully consider the ways in which the recommendations may go astray, in collaboration with experts and patients, to ensure that they are forces for good. In addition to the two papers I am directly involved with, there are several more pieces of work from Boise recommender system researchers and industry developers. In the poster session, you can see ‘Is Readability a Valuable Signal for Hashtag Recommendations?’ by Ion Madrazo and Sole Pera and ‘Genre Prediction to Inform the Recommendation Process’ by Nevena Dragovic and Sole. Sole and Nevena also have two demos with the relEVENT team at the RecTour 2016 workshop on recommenders for tourism: ‘Anything Fun Going On? A Simple Wizard to Avoid the Cold-Start Problem for Event Recommenders’ and ‘“One Size Doesn’t Fit All”: Helping Users Find Events from Multiple Perspectives’. We look forward to seeing you at RecSys. The Boise State University Dept. of Computer Science also just launched a new Ph.D program in computing, so if you are interested in doing graduate work in recommender systems (at either the Ph.D or M.S. level), talk to either Sole Pera or myself and we’d be happy to discuss the opportunities at Boise State. See you in Boston! Go Red Sox! And be sure to follow the conference on Twitter on the #recsys2016 hashtag.Listen to your Users
Don’t Hurt your Users
Other Boise RecSys Research
Join Us