Enhancing content-based recommendation with the task model of classification

Authors:
Yiwen Wang;Shenghui Wang;Natalia Stash;Lora Aroyo;Guus Schreiber
Affiliations:
Eindhoven University of Technology, Computer Science;VU University Amsterdam, Computer Science;Eindhoven University of Technology, Computer Science;Eindhoven University of Technology, Computer Science and VU University Amsterdam, Computer Science;VU University Amsterdam, Computer Science
Venue:
EKAW'10 Proceedings of the 17th international conference on Knowledge engineering and management by the masses
Year:
2010

Citing 5
Cited 2

Evaluating collaborative filtering recommender systems

ACM Transactions on Information Systems (TOIS)
Recommendations based on semantically enriched museum collections

Web Semantics: Science, Services and Agents on the World Wide Web
Knowledge engineering rediscovered: towards reasoning patterns for the semantic web

Proceedings of the fifth international conference on Knowledge capture
Semantic relations for content-based recommendations

Proceedings of the fifth international conference on Knowledge capture
An empirical study of instance-based ontology matching

ISWC'07/ASWC'07 Proceedings of the 6th international The semantic web and 2nd Asian conference on Asian semantic web conference

An approach for using complex event processing for adaptive augmented reality in cultural heritage domain: experience report

Proceedings of the 6th ACM International Conference on Distributed Event-Based Systems
Web supported emplotment: using object and event descriptions to facilitate storytelling online and in galleries

Proceedings of the 3rd Annual ACM Web Science Conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we define reusable inference steps for content-based recommender systems based on semantically-enriched collections. We show an instantiation in the case of recommending artworks and concepts based on a museum domain ontology and a user profile consisting of rated artworks and rated concepts. The recommendation task is split into four inference steps: realization, classification by concepts, classification by instances, and retrieval. Our approach is evaluated on real user rating data. We compare the results with the standard content-based recommendation strategy in terms of accuracy and discuss the added values of providing serendipitous recommendations and supporting more complete explanations for recommended items.