We examine the case of over-specialization in recommender systems, which results from returning items that are too similar to those previously rated by the user. We propose Outside-The-Box (otb) recommendation, which takes some risk to help users make fresh discoveries, while maintaining high relevance. The proposed formalization relies on item regions and attempts to identify regions that are under-exposed to the user. We develop a recommendation algorithm which achieves a compromise between relevance and risk to find otb items. We evaluate this approach on the MovieLens data set and compare our otb recommendations against conventional recommendation strategies.
%0 Conference Paper
%1 citeulike:7001315
%A Abbassi, Zeinab
%A Yahia, Sihem A.
%A Lakshmanan, Laks V. S.
%A Vassilvitskii, Sergei
%A Yu, Cong
%B RecSys '09: Proceedings of the third ACM conference on Recommender systems
%C New York, NY, USA
%D 2009
%I ACM
%K novelty recommender
%P 285--288
%R 10.1145/1639714.1639769
%T Getting recommender systems to think outside the box
%U http://dx.doi.org/10.1145/1639714.1639769
%X We examine the case of over-specialization in recommender systems, which results from returning items that are too similar to those previously rated by the user. We propose Outside-The-Box (otb) recommendation, which takes some risk to help users make fresh discoveries, while maintaining high relevance. The proposed formalization relies on item regions and attempts to identify regions that are under-exposed to the user. We develop a recommendation algorithm which achieves a compromise between relevance and risk to find otb items. We evaluate this approach on the MovieLens data set and compare our otb recommendations against conventional recommendation strategies.
%@ 978-1-60558-435-5
@inproceedings{citeulike:7001315,
abstract = {{We examine the case of over-specialization in recommender systems, which results from returning items that are too similar to those previously rated by the user. We propose Outside-The-Box (otb) recommendation, which takes some risk to help users make fresh discoveries, while maintaining high relevance. The proposed formalization relies on item regions and attempts to identify regions that are under-exposed to the user. We develop a recommendation algorithm which achieves a compromise between relevance and risk to find otb items. We evaluate this approach on the MovieLens data set and compare our otb recommendations against conventional recommendation strategies.}},
added-at = {2018-03-19T12:24:51.000+0100},
address = {New York, NY, USA},
author = {Abbassi, Zeinab and Yahia, Sihem A. and Lakshmanan, Laks V. S. and Vassilvitskii, Sergei and Yu, Cong},
biburl = {https://www.bibsonomy.org/bibtex/2a70d5b6dc225a932ad12f7ec77d65ce1/aho},
booktitle = {RecSys '09: Proceedings of the third ACM conference on Recommender systems},
citeulike-article-id = {7001315},
citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1639769},
citeulike-linkout-1 = {http://dx.doi.org/10.1145/1639714.1639769},
doi = {10.1145/1639714.1639769},
interhash = {4220a9e3b9b7a3c6a4f7d786dfe89190},
intrahash = {a70d5b6dc225a932ad12f7ec77d65ce1},
isbn = {978-1-60558-435-5},
keywords = {novelty recommender},
location = {New York, New York, USA},
pages = {285--288},
posted-at = {2010-04-20 18:55:30},
priority = {2},
publisher = {ACM},
timestamp = {2018-03-19T12:24:51.000+0100},
title = {{Getting recommender systems to think outside the box}},
url = {http://dx.doi.org/10.1145/1639714.1639769},
year = 2009
}