Abstract | ||
---|---|---|
We describe an artist recommendation system which inte- grates several heterogeneous data sources to form a holistic similarity space. Using social, semantic, and acoustic fea- tures, we learn a low-dimensional feature transformation which is optimized to reproduce human-derived measure- ments of subjective similarity between artists. By produc- ing low-dimensional representations of artists, our system is suitable for visualization and recommendation tasks. |
Year | Venue | Keywords |
---|---|---|
2009 | ISMIR 2013 | recommender system |
Field | DocType | Citations |
Recommender system,Feature transformation,Embedding,Information retrieval,Visualization,Computer science | Conference | 24 |
PageRank | References | Authors |
1.35 | 19 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Brian Mcfee | 1 | 440 | 24.05 |
Gert R. G. Lanckriet | 2 | 4769 | 296.98 |