eduzhai > Applied Sciences > Engineering >

Quick Lists Enriched Playlist Embeddings for Future Playlist Recommendation

  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: Recommending playlists to users in the context of a digital music service isa difficult task because a playlist is often more than the mere sum of itsparts. We present a novel method for generating playlist embeddings that areinvariant to playlist length and sensitive to local and global track ordering.The embeddings also capture information about playlist sequencing, and areenriched with side information about the playlist user. We show that theseembeddings are useful for generating next-best playlist recommendations, andthat side information can be used for the cold start problem.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...