publication . Conference object . Preprint . 2018

A hybrid approach to music playlist continuation based on playlist-song membership

Andreu Vall; Matthias Dorfer; Markus Schedl; Gerhard Widmer;
English
  • Published: 24 May 2018
  • Publisher: ACM Press
Abstract
Automated music playlist continuation is a common task of music recommender systems, that generally consists in providing a fitting extension to a given playlist. Collaborative filtering models, that extract abstract patterns from curated music playlists, tend to provide better playlist continuations than content-based approaches. However, pure collaborative filtering models have at least one of the following limitations: (1) they can only extend playlists profiled at training time; (2) they misrepresent songs that occur in very few playlists. We introduce a novel hybrid playlist continuation model based on what we name "playlist-song membership," that is, wheth...
Subjects
free text keywords: Computer Science - Information Retrieval, Information retrieval, Collaborative filtering, Continuation, Feature vector, Music information retrieval, Recommender system, Artificial neural network, Computer science
Funded by
EC| Con Espressione
Project
Con Espressione
Getting at the Heart of Things: Towards Expressivity-aware Computer Systems in Music
  • Funder: European Commission (EC)
  • Project Code: 670035
  • Funding stream: H2020 | ERC | ERC-ADG
41 references, page 1 of 3

[1] Gediminas Adomavicius and Alexander Tuzhilin. 2005. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering 17, 6 (June 2005), 734-749.

[2] Natalie Aizenberg, Yehuda Koren, and Oren Somekh. 2012. Build your own music recommender by modeling internet radio streams. In Proc. WWW. 1-10.

[3] Thierry Bertin-Mahieux, Daniel PW Ellis, Brian Whitman, and Paul Lamere. 2011. The million song dataset. In Proc. ISMIR. University of Miami, 591-596. [OpenAIRE]

[4] David M. Blei, Andrew Y. Ng, and Michael I. Jordan. 2003. Latent dirichlet allocation. Journal of Machine Learning Research 3 (2003), 993-1022.

[5] Geofray Bonnin and Dietmar Jannach. 2014. Automated generation of music playlists: Survey and experiments. Comput. Surveys 47, 2 (Nov. 2014), 1-35.

[6] Robin Burke. 2002. Hybrid recommender systems: Survey and experiments. User modeling and user-adapted interaction 12, 4 (2002), 331-370.

[7] Òscar Celma. 2010. Music recommendation and discovery. Springer. [OpenAIRE]

[8] Shuo Chen, Josh L. Moore, Douglas Turnbull, and Thorsten Joachims. 2012. Playlist prediction via metric embedding. In Proc. SIGKDD. 714-722.

[9] Sally Jo Cunningham, David Bainbridge, and Annette Falconer. 2006. “More of an art than a science”: Supporting the creation of playlists and mixes. In Proc. ISMIR.

[10] Najim Dehak, Patrick J Kenny, Réda Dehak, Pierre Dumouchel, and Pierre Ouellet. 2011. Front-end factor analysis for speaker verification. IEEE Transactions on Audio, Speech, and Language Processing 19, 4 (May 2011), 788-798.

[11] Sander Dieleman, Jan Schlüter, Colin Rafel, Eben Olson, Søren Kaae Sønderby, Daniel Nouri, Daniel Maturana, Martin Thoma, Eric Battenberg, Jack Kelly, Jeffrey De Fauw, Michael Heilman, Diogo Moitinho de Almeida, Brian McFee, Hendrik Weideman, Gábor Takács, Peter de Rivaz, Jon Crall, Gregory Sanders, Kashif Rasul, Cong Liu, Geofrey French, and Jonas Degrave. 2015. Lasagne: First release. (Aug. 2015). http://dx.doi.org/10.5281/zenodo.27878

[12] Hamid Eghbal-zadeh, Bernhard Lehner, Markus Schedl, and Gerhard Widmer. 2015. I-vectors for timbre-based music similarity and music artist classification. In Proc. ISMIR. [OpenAIRE]

[13] Hamid Eghbal-zadeh, Markus Schedl, and Gerhard Widmer. 2015. Timbral modeling for music artist recognition using i-vectors. In Proc. EUSIPCO. 1286-1290. [OpenAIRE]

[14] Arthur Flexer, Dominik Schnitzer, Martin Gasser, and Gerhard Widmer. 2008. Playlist generation using start and end songs. In Proc. ISMIR. 173-178. [OpenAIRE]

[15] Ben Frederickson. 2017. Fast python collaborative filtering for implicit datasets. (2017). https://github.com/benfred/implicit

41 references, page 1 of 3
Abstract
Automated music playlist continuation is a common task of music recommender systems, that generally consists in providing a fitting extension to a given playlist. Collaborative filtering models, that extract abstract patterns from curated music playlists, tend to provide better playlist continuations than content-based approaches. However, pure collaborative filtering models have at least one of the following limitations: (1) they can only extend playlists profiled at training time; (2) they misrepresent songs that occur in very few playlists. We introduce a novel hybrid playlist continuation model based on what we name "playlist-song membership," that is, wheth...
Subjects
free text keywords: Computer Science - Information Retrieval, Information retrieval, Collaborative filtering, Continuation, Feature vector, Music information retrieval, Recommender system, Artificial neural network, Computer science
Funded by
EC| Con Espressione
Project
Con Espressione
Getting at the Heart of Things: Towards Expressivity-aware Computer Systems in Music
  • Funder: European Commission (EC)
  • Project Code: 670035
  • Funding stream: H2020 | ERC | ERC-ADG
41 references, page 1 of 3

[1] Gediminas Adomavicius and Alexander Tuzhilin. 2005. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering 17, 6 (June 2005), 734-749.

[2] Natalie Aizenberg, Yehuda Koren, and Oren Somekh. 2012. Build your own music recommender by modeling internet radio streams. In Proc. WWW. 1-10.

[3] Thierry Bertin-Mahieux, Daniel PW Ellis, Brian Whitman, and Paul Lamere. 2011. The million song dataset. In Proc. ISMIR. University of Miami, 591-596. [OpenAIRE]

[4] David M. Blei, Andrew Y. Ng, and Michael I. Jordan. 2003. Latent dirichlet allocation. Journal of Machine Learning Research 3 (2003), 993-1022.

[5] Geofray Bonnin and Dietmar Jannach. 2014. Automated generation of music playlists: Survey and experiments. Comput. Surveys 47, 2 (Nov. 2014), 1-35.

[6] Robin Burke. 2002. Hybrid recommender systems: Survey and experiments. User modeling and user-adapted interaction 12, 4 (2002), 331-370.

[7] Òscar Celma. 2010. Music recommendation and discovery. Springer. [OpenAIRE]

[8] Shuo Chen, Josh L. Moore, Douglas Turnbull, and Thorsten Joachims. 2012. Playlist prediction via metric embedding. In Proc. SIGKDD. 714-722.

[9] Sally Jo Cunningham, David Bainbridge, and Annette Falconer. 2006. “More of an art than a science”: Supporting the creation of playlists and mixes. In Proc. ISMIR.

[10] Najim Dehak, Patrick J Kenny, Réda Dehak, Pierre Dumouchel, and Pierre Ouellet. 2011. Front-end factor analysis for speaker verification. IEEE Transactions on Audio, Speech, and Language Processing 19, 4 (May 2011), 788-798.

[11] Sander Dieleman, Jan Schlüter, Colin Rafel, Eben Olson, Søren Kaae Sønderby, Daniel Nouri, Daniel Maturana, Martin Thoma, Eric Battenberg, Jack Kelly, Jeffrey De Fauw, Michael Heilman, Diogo Moitinho de Almeida, Brian McFee, Hendrik Weideman, Gábor Takács, Peter de Rivaz, Jon Crall, Gregory Sanders, Kashif Rasul, Cong Liu, Geofrey French, and Jonas Degrave. 2015. Lasagne: First release. (Aug. 2015). http://dx.doi.org/10.5281/zenodo.27878

[12] Hamid Eghbal-zadeh, Bernhard Lehner, Markus Schedl, and Gerhard Widmer. 2015. I-vectors for timbre-based music similarity and music artist classification. In Proc. ISMIR. [OpenAIRE]

[13] Hamid Eghbal-zadeh, Markus Schedl, and Gerhard Widmer. 2015. Timbral modeling for music artist recognition using i-vectors. In Proc. EUSIPCO. 1286-1290. [OpenAIRE]

[14] Arthur Flexer, Dominik Schnitzer, Martin Gasser, and Gerhard Widmer. 2008. Playlist generation using start and end songs. In Proc. ISMIR. 173-178. [OpenAIRE]

[15] Ben Frederickson. 2017. Fast python collaborative filtering for implicit datasets. (2017). https://github.com/benfred/implicit

41 references, page 1 of 3
Powered by OpenAIRE Research Graph
Any information missing or wrong?Report an Issue