Understanding user preferences and goals in recommender systems Martijn C. Willemsen Eindhoven University of Technology, Eindhoven, the Netherlands M.C.Willemsen@tue.nl www.martijnwillemsen.nl Abstract Recommender systems typically use collaborative filtering: information from your preferences (i.e. your ratings) is combined with that of other users to pre- dict what other items you might also like. Much of the research in the field has focused on building algorithms that provide recommendations based purely on predicted accuracy [5]. However, these models make strong assumptions about how preferences come about, how stable they are, and how they can be mea- sured [4]. Having a background in decision psychology I have studied how the preference elicitation methods of recommender systems can be better under- stood and improved based on psychological insights. I will illustrate this with an example of new choice-based preference interfaces we have developed. Users are more satisfied with a method that measures their preferences through a series of choices than with a rating-based preference elicitation, because the rating-based is more effortful and provides more obscure movies [2]. However, a drawback is that recommendation lists of choice-based preference elicitation contain mostly popular movies, and further research has investigated that showing trailers can help to reduce this popularity effect a bit as users are able to use the trailer to inspect less well-known items [3]. Moreover, recommender systems should also align with user goals. Many real- life recommender systems are evaluated mostly on (implicit) behavioral data such as clicks streams and viewing times. However, such an approach has limitations and I will show how a user-centric approach can help better understand why users are satisfied or not, for example why users prefer diversify over prediction accuracy as it reduces choice difficulty [8]. The behaviorist approach to evalua- tion also misses that users’ short term goals (i.e. their current behavior) might not be representative of the goals they want to attain (i.e. their desired behav- ior) [1]. This is especially relevant in health and life style domains [6] where people are in need of support while changing their current behavior. I will elab- orate on an example in the energy recommendation domain, and show how a different type of recommender approach and interface might help users to save more energy [7]. 2 Martijn C. Willemsen Speaker Dr. Martijn Willemsen is an expert on human decision making in interac- tive systems. He is working as an associate professor in the Human-Technology Interaction group of Eindhoven University of Technology (The Netherlands). His primary interests lie in the understanding of cognitive processes of decision making by means of process tracing and in the application of decision making theory in interactive systems such as recommender systems. He is also an expert on user-centric evaluation of adaptive systems. He is part of the core team of the Customer Journey Research Program in the Data Science Center Eindhoven (DSC/e) and is teaching in the joint BSc and MSc data science programs of the Jheronimus Academy of Data Science (jads.nl). References 1. M. D. Ekstrand and M. C. Willemsen. Behaviorism is Not Enough: Better Recom- mendations Through Listening to Users. In Proceedings of the 10th ACM Conference on Recommender Systems, RecSys ’16, pages 221–224, New York, NY, USA, 2016. ACM. 2. M. P. Graus and M. C. Willemsen. Improving the User Experience During Cold Start Through Choice-Based Preference Elicitation. In Proceedings of the 9th ACM Conference on Recommender Systems, RecSys ’15, pages 273–276, New York, NY, USA, 2015. ACM. 3. Graus, Mark P. and Willemsen, Martijn C. Can Trailers Help to Alleviate Popularity Bias in Choice-Based Preference Elicitation? In Proceedings of the Joint Workshop on Interfaces and Human Decision Making for Recommender Systems, volume 1679, Boston, MA, 2016. CEUR Workshop Proceedings. 4. A. Jameson, M. C. Willemsen, A. Felfernig, M. d. Gemmis, P. Lops, G. Semeraro, and L. Chen. Human Decision Making and Recommender Systems. In F. Ricci, L. Rokach, and B. Shapira, editors, Recommender Systems Handbook, pages 611– 648. Springer US, 2015. DOI: 10.1007/978-1-4899-7637-6 18. 5. S. M. McNee, J. Riedl, and J. A. Konstan. Being accurate is not enough: how accuracy metrics have hurt recommender systems. In CHI ’06 extended abstracts on Human factors in computing systems, CHI EA ’06, pages 1097–1101, Montreal, Quebec, Canada, 2006. ACM. ACM ID: 1125659. 6. M. Radha, M. C. Willemsen, M. Boerhof, and W. A. IJsselsteijn. Lifestyle Recom- mendations for Hypertension Through Rasch-based Feasibility Modeling. In Pro- ceedings of the 2016 Conference on User Modeling Adaptation and Personalization, UMAP ’16, pages 239–247, New York, NY, USA, 2016. ACM. 7. Starke, Alain, M. C. Willemsen, and Snijders, Chris. Effective user interface designs to increase energy-efficient behavior in a rasch-based energy recommender system. In Proceedings of the 11th ACM Conference on Recommender Systems, RecSys ’17, New York, NY, USA, 2017. ACM. 8. M. C. Willemsen, M. P. Graus, and B. P. Knijnenburg. Understanding the role of latent feature diversification on choice difficulty and satisfaction. User Modeling and User-Adapted Interaction, 26(4):347–389, Oct. 2016.