Offline and Online Evaluation of News Recommender Systems at swissinfo.ch

被引:92
作者
Garcin, Florent [1 ]
Faltings, Boi [1 ]
Donatsch, Olivier [2 ]
Alazzawi, Ayar [2 ]
Bruttin, Christophe [2 ]
Huber, Amr [2 ]
机构
[1] Ecole Polytech Fed Lausanne, Artificial Intelligence Lab, Lausanne, Switzerland
[2] Swiss Broadcasting Corp, SWI Swissinfo Ch, Bern, Switzerland
来源
PROCEEDINGS OF THE 8TH ACM CONFERENCE ON RECOMMENDER SYSTEMS (RECSYS'14) | 2014年
关键词
recommender system; news; real-time; live; evaluation;
D O I
10.1145/2645710.2645745
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We report on the live evaluation of various news recommender systems conducted on the website swissinfo.ch. We demonstrate that there is a major difference between offline and online accuracy evaluations. In an offline setting, recommending most popular stories is the best strategy, while in a live environment this strategy is the poorest. For online setting, context-tree recommender systems which profile the users in real-time improve the click-through rate by up to 35%. The visit length also increases by a factor of 2.5. Our experience holds important lessons for the evaluation of recommender systems with offline data as well as for the use of the click-through rate as a performance indicator.
引用
收藏
页码:169 / 176
页数:8
相关论文
共 23 条
[1]  
Ahn J. W., 2007, P 16 INT C WORLD WID, P11, DOI [10.1145/1242572.1242575, DOI 10.1145/1242572.1242575]
[2]  
[Anonymous], 2010, EDBT ICDT WORKSH 10
[3]  
[Anonymous], 2013, P 2013 WORKSHOP LIVI
[4]  
[Anonymous], 2010, Proceedings of the 19th International Conference on World Wide Web, WWW'10, page, DOI DOI 10.1145/1772690.1772754
[5]  
[Anonymous], 2007, Google news personalization: scalable online collaborative filtering, DOI DOI 10.1145/1242572.1242610
[6]  
[Anonymous], 2013, P 2013 INT NEWS RECO
[7]  
Billsus D, 1999, CISM COUR L, P99
[8]   Hybrid recommender systems: Survey and experiments [J].
Burke, R .
USER MODELING AND USER-ADAPTED INTERACTION, 2002, 12 (04) :331-370
[9]  
Chen L, 2010, LECT NOTES COMPUT SC, V6075, P375, DOI 10.1007/978-3-642-13470-8_35
[10]  
Cremonesi P, 2010, P 4 ACM C REC SYST, P39, DOI DOI 10.1145/1864708.1864721