Beitrag in einem Tagungsband
A Comparison of content-based Tag Recommendations in Folksonomy Systems
Details zur Publikation
Autor(inn)en: | Jäschke, R.; Stumme, G.; Hotho, A.; Illig, J. |
Herausgeber: | Wolff Erich Karl, Palchunov E. Dmitry, Zagoruiko G. Nikolay, Andelfinger Urs |
Verlag: | Springer |
Verlagsort / Veröffentlichungsort: | Berlin/Heidelberg |
Publikationsjahr: | 2011 |
Seitenbereich: | 136-149 |
Buchtitel: | Postproceedings of the International Conference on Knowledge Processing in Practice (KPP 2007) |
Titel der Buchreihe: | Lecture Notes in Computer Science |
Zusammenfassung, Abstract
Recommendation algorithms and multi-class classifiers can supportusers of social bookmarking systems in assigning tags to theirbookmarks. Content based recommenders are the usual approach forfacing the cold start problem, i.e., when a bookmark is uploaded forthe first time and no information from other users can be exploited.In this paper, we evaluate several recommendation algorithms in acold-start scenario on a large real-world dataset.
Recommendation algorithms and multi-class classifiers can supportusers of social bookmarking systems in assigning tags to theirbookmarks. Content based recommenders are the usual approach forfacing the cold start problem, i.e., when a bookmark is uploaded forthe first time and no information from other users can be exploited.In this paper, we evaluate several recommendation algorithms in acold-start scenario on a large real-world dataset.
Schlagwörter
folksonomy, recommender, tag
