Does Wikipedia Information Help Netflix Predictions?

Author

Lees-Miller, John ; Anderson, Fraser ; Hoehn, Bret ; Greiner, Russell

Author_Institution

Dept. of Comput. Sci., Univ. of Alberta, AB

fYear

2008

fDate

11-13 Dec. 2008

Firstpage

337

Lastpage

343

Abstract

We explore several ways to estimate movie similarity from the free encyclopedia Wikipedia with the goal of improving our predictions for the Netflix Prize. Our system first uses the content and hyperlink structure of Wikipedia articles to identify similarities between movies. We then predict a user´s unknown ratings by using these similarities in conjunction with the user´s known ratings to initialize matrix factorization and k-Nearest Neighbours algorithms. We blend these results with existing ratings-based predictors. Finally, we discuss our empirical results, which suggest that external Wikipedia data does not significantly improve the overall prediction accuracy.

Keywords

information networks; matrix algebra; search engines; Netflix Prize; Wikipedia articles; free encyclopedia; hyperlink structure; k-nearest neighbours algorithms; matrix factorization; Accuracy; Collaboration; Encyclopedias; Information filtering; Information filters; Internet; Motion pictures; Probes; Voting; Wikipedia; collaborative filtering; hybrid; netflix; recommender system; wikipedia;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Applications, 2008. ICMLA '08. Seventh International Conference on

Conference_Location

San Diego, CA

Print_ISBN

978-0-7695-3495-4

Type

conf

DOI

10.1109/ICMLA.2008.121

Filename

4724995