On Weighted Hybrid Track Recommendations
Music is a highly subjective domain, which makes it a challenging research area for recommender systems. In this paper, we present our TRecS (Track Recommender System) prototype, a hybrid recommender that blends three different recommender techniques into one score. Since traceability is an important issue for the acceptance of recommender systems by users, we have implemented a detailed explanation feature that supports transparency about the contribution of each sub-recommender for the overall result. To avoid overspecialization, TRecS peppers the result list with recommendations that are based on a serendipity metric. This way, users can benefit from both recommendations aligned with their current taste while gaining some diversification.
Unable to display preview. Download preview PDF.
- 1.Zhang, Y.C., Séaghdha, D.Ó., Quercia, D., Jambor, T.: Auralist: Introducing Serendipity Into Music Recommendation. In: Adar, E., Teevan, J., Agichtein, E., Maarek, Y. (eds.) WSDM, pp. 13–22. ACM (2012)Google Scholar
- 4.Baeza-Yates, R.A., Ribeiro-Neto, B.A.: Modern Information Retrieval - The Concepts and Technology Behind Search, 2nd edn. Pearson Education Ltd., Harlow (2011)Google Scholar
- 5.Ziegler, C.N., McNee, S., Konstan, J., Lausen, G.: Improving Recommendation Lists Through Topic Diversification. In: Proceedings of the 14th International World Wide Web Conference, Chiba, Japan. ACM Press (May 2005)Google Scholar