TY - GEN
T1 - Recommendations with a purpose
AU - Jannach, Dietmar
AU - Adomavicius, Gediminas
PY - 2016/9/7
Y1 - 2016/9/7
N2 - The purpose of recommenders is often summarized as "help the users find relevant items", and the predominant opera- tionalization of this goal has been to focus on the ability to numerically estimate the users' preferences for unseen items or to provide users with item lists ranked in accordance to the estimated preferences. This dominant, albeit narrow, view of the recommendation problem has been tremendously helpful in advancing research in diferent ways, e.g., through the establishment of standardized evaluation procedures and metrics. In reality, recommender systems can serve a vari- ety of purposes from the point of view of both consumers and providers. Most of the purposes, however, are signif- icantly underexplored, even though many of them are ar- guably more aligned with the real-world expectations for rec- ommenders than our current predominant paradigm. There- fore, it is important to revisit our conceptualizations of the potential goals of recommenders and their operationalization as research problems. In this paper, we discuss a framework of recommendation goals and purposes and highlight pos- sible future directions and challenges related to the opera- tionalization of such alternative problem formulations.
AB - The purpose of recommenders is often summarized as "help the users find relevant items", and the predominant opera- tionalization of this goal has been to focus on the ability to numerically estimate the users' preferences for unseen items or to provide users with item lists ranked in accordance to the estimated preferences. This dominant, albeit narrow, view of the recommendation problem has been tremendously helpful in advancing research in diferent ways, e.g., through the establishment of standardized evaluation procedures and metrics. In reality, recommender systems can serve a vari- ety of purposes from the point of view of both consumers and providers. Most of the purposes, however, are signif- icantly underexplored, even though many of them are ar- guably more aligned with the real-world expectations for rec- ommenders than our current predominant paradigm. There- fore, it is important to revisit our conceptualizations of the potential goals of recommenders and their operationalization as research problems. In this paper, we discuss a framework of recommendation goals and purposes and highlight pos- sible future directions and challenges related to the opera- tionalization of such alternative problem formulations.
KW - Foundations of Recommender Systems
KW - Recommendation Goals and Purposes
UR - http://www.scopus.com/inward/record.url?scp=84991270596&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84991270596&partnerID=8YFLogxK
U2 - 10.1145/2959100.2959186
DO - 10.1145/2959100.2959186
M3 - Conference contribution
AN - SCOPUS:84991270596
T3 - RecSys 2016 - Proceedings of the 10th ACM Conference on Recommender Systems
SP - 7
EP - 10
BT - RecSys 2016 - Proceedings of the 10th ACM Conference on Recommender Systems
PB - Association for Computing Machinery, Inc
T2 - 10th ACM Conference on Recommender Systems, RecSys 2016
Y2 - 15 September 2016 through 19 September 2016
ER -