Name
Abstract | ||
|---|---|---|
Our goal is to establish a method for predicting users' potential preference. We define a potential preference as a preference for the unknown genres for the target user. However, it is difficult to predict the potential preference by conventional recommender systems because there is little or no preference data (i.e. ratings for items) for the users' unknown genres. Accordingly, we propose a collaborative filtering for predicting the users' potential preference by their ratings in their known genres. Experimental results using MovieLens data sets showed that the genre relevance influences the prediction accuracy of the potential preference in the unknown genres. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1007/978-3-642-23866-6_5 | KES (4) |
Keywords | Field | DocType |
preference data,target user,known genre,genre relevance,movielens data set,prediction accuracy,potential preference,conventional recommender system,unknown genre,collaborative filtering | Recommender system,Data set,World Wide Web,Collaborative filtering,Information retrieval,Computer science,MovieLens,Preference learning | Conference |
Citations | PageRank | References |
1 | 0.39 | 9 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kenta Oku | 1 | 84 | 14.81 |
| Ta Son Tung | 2 | 1 | 0.39 |
| Fumio Hattori | 3 | 164 | 26.81 |