Integrating semantic relatedness in a collaborative filtering system
| dc.contributor.author | Ferrara, Felice | en |
| dc.contributor.author | Tasso, Carlo | en |
| dc.contributor.editor | Augstein, Mirjam | en |
| dc.contributor.editor | Heckmann, Dominik | en |
| dc.contributor.editor | Herder, Eelco | en |
| dc.date.accessioned | 2017-11-15T15:01:28Z | |
| dc.date.available | 2017-11-15T15:01:28Z | |
| dc.date.issued | 2012 | |
| dc.description.abstract | Collaborative Filtering (CF) recommender systems use opinions of people for filtering relevant information. The accuracy of these applications depends on the mechanism used to filter and combine the opinions (the feedback) provided by users. In this paper we propose a mechanism aimed at using semantic relations extracted from Wikipedia in order to adaptively filter and combine the feedback of people. The semantic relatedness among the concepts/pages of Wikipedia is used to identify the opinions which are more significant for predicting a rating for an item. We show that our approach improves the accuracy of the predictions and it also opens opportunities for providing explanations on the obtained recommendations. | en |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/5117 | |
| dc.language.iso | en | en |
| dc.relation.ispartof | ABIS 2012 | de_DE |
| dc.subject | Collaborative Filtering (CF) | en |
| dc.title | Integrating semantic relatedness in a collaborative filtering system | en |
| dc.type | Text/Conference Paper | de_DE |
| gi.citation.publisherPlace | Konstanz | de_DE |
| gi.document.quality | digidoc | en |