A content boosted collaborative filtering approach for movie recommendation based on local & global similarity and missing data prediction

Özbal G., Karaman H., ALPASLAN F. N.

25th International Symposium on Computer and Information Sciences, ISCIS 2010, London, England, 22 - 24 September 2010, vol.62 LNEE, pp.109-112, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 62 LNEE
  • Doi Number: 10.1007/978-90-481-9794-1_22
  • City: London
  • Country: England
  • Page Numbers: pp.109-112
  • Keywords: Collaborative Filtering, Floyd Warshall Algorithm, Pearson Correlation Coefficient, Recommender Systems
  • Middle East Technical University Affiliated: Yes

Abstract

Many recommender systems lack in accuracy when the data used throughout the recommendation process is sparse. Our study addresses this limitation by means of a content boosted collaborative filtering approach applied to the task of movie recommendation. We combine two different approaches previously proved to be successful individually and improve over them by processing the content information of movies, as confirmed by our empirical evaluation results. © 2011 Springer Science+Business Media B.V.