Using Linked Open Data for Novel Artist Recommendations

Stephan Baumann; Rafael Schirru

In: Fabien Gouyon; Perfecto Herrera; Luis Gustavo Martins; Meinard Müller (Hrsg.). Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. International Conference on Music Information Retrieval (ISMIR-2012), October 8-12, Porto, Portugal, FEUP Edições, 2012.


As part of the Linking Open Data initiative many (community) platforms have made their data freely available in recent years. In these systems valuable information from the music domain can be found. We use this data for contentbased music recommendations thus not requiring expensive metadata annotations from domain experts. In this demo paper we present a system that uses data from the Freebase platform to describe music artists. The artists are represented using metadata profiles that characterize different aspects of their works such as genres and instrumentations but also collaborations with other artists. The system uses a Lucene index for content-based artist recommendations aiming at Web-scalability.


Weitere Links

LBD1.pdf (pdf, 305 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence