The LAST.FM dataset corresponds to user-artist plays on Last.fm online music system released during HETRec 2011 Workshop. It contains social networking, tagging, and music artists listening information from a set of 2K users. LibraryThing represents books' ratings collected in the LibraryThing website\footnote{\url{https://www.librarything.com/}} community. It contains social networking, tagging and rating information on a [1..10] scale. Yahoo Movies (Yahoo! Webscope dataset ydata-ymovies-user-movie-ratings-content-v1_0) contains movies ratings generated on Yahoo! Movies up to November 2003. It provides content, demographic and ratings information on a [1..5] scale, and mappings to \movielens and EachMovie datasets. Facebook Movies, Facebook Music and Facebook Books datasets have been released for the Linked Open Data challenge co-located with ESWC 2015, and they refer to movies, music and books domains respectively. Only implicit feedback is available for these datasets, but for each item a link to DBpedia is provided. In order to map items in LAST.FM and LibraryThing to DBpedia resources, we exploited a freely available mapping. For the remaining one (Yahoo Movies), we extracted all the updated items-features mappings (Yahoo Movies, LibraryThing,LAST.FM, Facebook Movies, Facebook Music and Facebook Books).
Datasets statistics are shown in Table.
This algorithm has been developed by Vito Walter Anelli and Joseph Trotta while working at SisInf Lab under the supervision of Tommaso Di Noia.
Tommaso Di Noia, tommaso [dot] dinoia [at] poliba [dot] it
Vito Walter Anelli, vitowalter [dot] anelli [at] poliba [dot] it
Joseph Trotta, joseph [dot] trotta [at] poliba [dot] it