%0 Conference Paper %1 Schnabel_2020 %A Schnabel, Tobias %A Ramos, Gonzalo %A Amershi, Saleema %B Fourteenth ACM Conference on Recommender Systems %D 2020 %I ACM %K interface preference-elicitation recommender recsys2020 user-centered-design %P 398-407 %R 10.1145/3383313.3412267 %T “Who doesn't like dinosaurs?” Finding and Eliciting Richer Preferences for Recommendation %U https://doi.org/10.1145%2F3383313.3412267 %X Real-world recommender systems often allow users to adjust the presented content through a variety of preference elicitation techniques such as “liking” or interest profiles. These elicitation techniques trade-off time and effort to users with the richness of the signal they provide to learning component driving the recommendations. In this paper, we explore this trade-off, seeking new ways for people to express their preferences with the goal of improving communication channels between users and the recommender system. Through a need-finding study, we observe the patterns in how people express their preferences during curation task, propose a taxonomy for organizing them, and point out research opportunities. We present a case study that illustrates how using this taxonomy to design an onboarding experience can lead to more accurate machine-learned recommendations while maintaining user satisfaction under low effort.

@inproceedings{Schnabel_2020, abstract = {Real-world recommender systems often allow users to adjust the presented content through a variety of preference elicitation techniques such as “liking” or interest profiles. These elicitation techniques trade-off time and effort to users with the richness of the signal they provide to learning component driving the recommendations. In this paper, we explore this trade-off, seeking new ways for people to express their preferences with the goal of improving communication channels between users and the recommender system. Through a need-finding study, we observe the patterns in how people express their preferences during curation task, propose a taxonomy for organizing them, and point out research opportunities. We present a case study that illustrates how using this taxonomy to design an onboarding experience can lead to more accurate machine-learned recommendations while maintaining user satisfaction under low effort.}, added-at = {2020-11-28T18:18:54.000+0100}, author = {Schnabel, Tobias and Ramos, Gonzalo and Amershi, Saleema}, biburl = {https://www.bibsonomy.org/bibtex/20f940aab41bbcfc806276b583c9d82ff/brusilovsky}, booktitle = {Fourteenth {ACM} Conference on Recommender Systems}, description = {“Who doesn’t like dinosaurs?” Finding and Eliciting Richer Preferences for Recommendation | Fourteenth ACM Conference on Recommender Systems}, doi = {10.1145/3383313.3412267}, interhash = {c233fd3cb8333561577b5c62ce399860}, intrahash = {0f940aab41bbcfc806276b583c9d82ff}, keywords = {interface preference-elicitation recommender recsys2020 user-centered-design}, month = sep, pages = {398-407}, publisher = {{ACM}}, timestamp = {2020-11-28T18:19:19.000+0100}, title = {{\textquotedblleft}Who doesn't like dinosaurs?{\textquotedblright} Finding and Eliciting Richer Preferences for Recommendation}, url = {https://doi.org/10.1145%2F3383313.3412267}, year = 2020 }