On Unexpectedness in Recommender Systems: Or How to Expect the Unexpected
P. Adamopoulos, und A. Tuzhilin. DiveRS 2011 – ACM RecSys 2011 Workshop on Novelty and Diversity in Recommender Systems, New York, NY, USA, ACM, (Oktober 2011)
Zusammenfassung
Although the broad social and business success of recommender systems has been achieved across several domains, there is still a long way to go in terms of user satisfaction. One of the key dimensions for improvement is the concept of unexpectedness. In this paper, we propose a model to improve user satisfaction by
generating unexpected recommendations based on the utility theory of economics. In particular, we propose a new concept of
unexpectedness as recommending to users those items that depart from what they expect from the system. We define and formalize the concept of unexpectedness and discuss how it differs from the
related notions of novelty, serendipity and diversity. We also measure the quality of recommendations using specific metrics
under certain utility functions. Finally, we provide unexpected recommendations of high quality and conduct several experiments
on a real-world dataset to compare our recommendation results with some other standard baseline methods. Our proposed
approach outperforms these baseline methods in terms of unexpectedness while avoiding accuracy loss.
%0 Conference Paper
%1 adamopoulos2011
%A Adamopoulos, Panagiotis
%A Tuzhilin, Alexander
%B DiveRS 2011 – ACM RecSys 2011 Workshop on Novelty and Diversity in Recommender Systems
%C New York, NY, USA
%D 2011
%I ACM
%K Evaluation Recommendations Recommender Serendipity Systems Theory Unexpectedness Utility myown
%T On Unexpectedness in Recommender Systems: Or How to Expect the Unexpected
%U http://ceur-ws.org/Vol-816/paper2.pdf
%X Although the broad social and business success of recommender systems has been achieved across several domains, there is still a long way to go in terms of user satisfaction. One of the key dimensions for improvement is the concept of unexpectedness. In this paper, we propose a model to improve user satisfaction by
generating unexpected recommendations based on the utility theory of economics. In particular, we propose a new concept of
unexpectedness as recommending to users those items that depart from what they expect from the system. We define and formalize the concept of unexpectedness and discuss how it differs from the
related notions of novelty, serendipity and diversity. We also measure the quality of recommendations using specific metrics
under certain utility functions. Finally, we provide unexpected recommendations of high quality and conduct several experiments
on a real-world dataset to compare our recommendation results with some other standard baseline methods. Our proposed
approach outperforms these baseline methods in terms of unexpectedness while avoiding accuracy loss.
@inproceedings{adamopoulos2011,
abstract = {Although the broad social and business success of recommender systems has been achieved across several domains, there is still a long way to go in terms of user satisfaction. One of the key dimensions for improvement is the concept of unexpectedness. In this paper, we propose a model to improve user satisfaction by
generating unexpected recommendations based on the utility theory of economics. In particular, we propose a new concept of
unexpectedness as recommending to users those items that depart from what they expect from the system. We define and formalize the concept of unexpectedness and discuss how it differs from the
related notions of novelty, serendipity and diversity. We also measure the quality of recommendations using specific metrics
under certain utility functions. Finally, we provide unexpected recommendations of high quality and conduct several experiments
on a real-world dataset to compare our recommendation results with some other standard baseline methods. Our proposed
approach outperforms these baseline methods in terms of unexpectedness while avoiding accuracy loss.},
added-at = {2012-09-28T17:57:34.000+0200},
address = {New York, NY, USA},
author = {Adamopoulos, Panagiotis and Tuzhilin, Alexander},
biburl = {https://www.bibsonomy.org/bibtex/21cb735bcda3fb3bf7663aa4ff03e6573/padamop},
booktitle = {DiveRS 2011 – ACM RecSys 2011 Workshop on Novelty and Diversity in Recommender Systems},
citeulike-article-id = {10312220},
citeulike-linkout-0 = {http://ceur-ws.org/Vol-816/paper2.pdf},
interhash = {9937cc83d3352d55642364fe09c63df3},
intrahash = {1cb735bcda3fb3bf7663aa4ff03e6573},
keywords = {Evaluation Recommendations Recommender Serendipity Systems Theory Unexpectedness Utility myown},
location = {Chicago, IL, USA},
month = {October},
posted-at = {2012-02-03 19:28:21},
publisher = {ACM},
series = {RecSys 2011},
timestamp = {2012-09-28T17:57:34.000+0200},
title = {On Unexpectedness in Recommender Systems: Or How to Expect the Unexpected},
url = {http://ceur-ws.org/Vol-816/paper2.pdf},
year = 2011
}