%0 Conference Paper %1 gadanho2007addressing %A Gadanho, Sandra Clara %A Lhuillier, Nicolas %B Proceedings of the 2007 ACM conference on Recommender systems %C New York, NY, USA %D 2007 %I ACM %K implicit-feedback recommendation search social-search study %P 97--104 %R 10.1145/1297231.1297248 %T Addressing uncertainty in implicit preferences %U http://doi.acm.org/10.1145/1297231.1297248 %X The increasing amount of content available via digital television has made TV program recommenders valuable tools. In order to provide personalized recommendations, recommender systems need to collect information about user preferences. Since users are reluctant to invest much time in explicitly expressing their interests, preferences often need to be implicitly inferred through data gathered by monitoring user behavior. Which is, alas, less reliable.</p> <p>This article addresses the problem of learning TV preferences based on tracking the programs users have watched, whilst dealing with the varying degrees of reliability in such information. Three approaches to the problem are discussed: use all information equally; weight information by its reliability or simply discard the most unreliable information.</p> <p>Experimental results for these three approaches are presented and compared using a content-based filtering recommender built on a Na&#239;ve Bayes classifier. %@ 978-1-59593-730--8

@inproceedings{gadanho2007addressing, abstract = {The increasing amount of content available via digital television has made TV program recommenders valuable tools. In order to provide personalized recommendations, recommender systems need to collect information about user preferences. Since users are reluctant to invest much time in explicitly expressing their interests, preferences often need to be implicitly inferred through data gathered by monitoring user behavior. Which is, alas, less reliable.</p> <p>This article addresses the problem of learning TV preferences based on tracking the programs users have watched, whilst dealing with the varying degrees of reliability in such information. Three approaches to the problem are discussed: use all information equally; weight information by its reliability or simply discard the most unreliable information.</p> <p>Experimental results for these three approaches are presented and compared using a content-based filtering recommender built on a Na&#239;ve Bayes classifier.}, acmid = {1297248}, added-at = {2011-07-21T16:44:36.000+0200}, address = {New York, NY, USA}, author = {Gadanho, Sandra Clara and Lhuillier, Nicolas}, biburl = {https://www.bibsonomy.org/bibtex/2f3d3929aef97d90ef55dd6dfe060b525/beate}, booktitle = {Proceedings of the 2007 ACM conference on Recommender systems}, description = {Addressing uncertainty in implicit preferences}, doi = {10.1145/1297231.1297248}, interhash = {82f451191a2913e1bbf23ba08a262e5a}, intrahash = {f3d3929aef97d90ef55dd6dfe060b525}, isbn = {978-1-59593-730--8}, keywords = {implicit-feedback recommendation search social-search study}, location = {Minneapolis, MN, USA}, numpages = {8}, pages = {97--104}, publisher = {ACM}, series = {RecSys '07}, timestamp = {2011-07-21T16:44:36.000+0200}, title = {Addressing uncertainty in implicit preferences}, url = {http://doi.acm.org/10.1145/1297231.1297248}, year = 2007 }