ABSTRACT
Playlist generation is a special form of music recommendation where the problem is to create a sequence of tracks to be played next, given a number of seed tracks. In academia, the evaluation of playlisting techniques is often done by assessing with the help of information retrieval measures if an algorithm is capable of selecting those tracks that also a human would pick next. Such approaches however cannot capture other factors, e.g., the homogeneity of the tracks that can determine the quality perception of playlists. In this work, we report the results of a multi-metric comparison of different academic approaches and a commercial playlisting service. Our results show that all tested techniques generate playlists with certain biases, e.g., towards very popular tracks, and often create playlists continuations that are quite different from those that are created by real users.
- P. Adamopoulos and A. Tuzhilin. On Over-specialization and Concentration Bias of Recommendations:Probabilistic Neighborhood Selection in Collaborative Filtering Systems. In Proc. RecSys '14, pages 153--160, 2014. Google ScholarDigital Library
- A. Andric and G. Haus. Automatic Playlist Generation based on Tracking User's ListeningHabits. Multimedia Tools and Applications, 29(2):127--151, 2006. Google ScholarDigital Library
- J.-J. Aucouturier and F. Pachet. Scaling Up Music Playlist Generation. In Proc. ICME, volume 1, pages 105--108, 2002.Google ScholarCross Ref
- W. Balkema and F. van der Heijden. Music Playlist Generation by Assimilating GMMs into SOMs. Pattern Recognition Letters, 31(11):1396--1402, 2010. Google ScholarDigital Library
- G. Bonnin and D. Jannach. Evaluating the Quality of Playlists Based on Hand-Crafted Samples. In Proc. ISMIR, pages 263--268, 2013.Google Scholar
- G. Bonnin and D. Jannach. Automated Generation of Music Playlists: Survey and Experiments. ACM Comput. Surv., 47(2):26:1--26:35, 2014. Google ScholarDigital Library
- S. Chen, J. L. Moore, D. Turnbull, and T. Joachims. Playlist Prediction via Metric Embedding. In Proc. KDD, pages 714--722, 2012. Google ScholarDigital Library
- C. Desrosiers and G. Karypis. A Comprehensive Survey of Neighborhood-based RecommendationMethods. In RSs Handbook, pages 107--144. Springer US, 2011.Google Scholar
- N. Hariri, B. Mobasher, and R. Burke. Context-Aware Music Recommendation Based on Latent Topic SequentialPatterns. In Proc. RecSys, pages 131--138, 2012. Google ScholarDigital Library
- T. Jambor and J. Wang. Optimizing Multiple Objectives in Collaborative Filtering. In Proc. RecSys '10, pages 55--62, 2010. Google ScholarDigital Library
- D. Jannach, L. Lerche, and I. Kamehkhosh. Beyond "Hitting the Hits": Generating Coherent Music PlaylistContinuations with the Right Tracks. In Proc. RecSys, pages 187--194, 2015. Google ScholarDigital Library
- D. Jannach, L. Lerche, I. Kamehkhosh, and M. Jugovac. What Recommenders Recommend: An Analysis of Recommendation Biasesand Possible Countermeasures. User Modeling and User-Adapted Interaction, pages 1--65, 2015. Google ScholarDigital Library
- M. Kamalzadeh, D. Baur, and T. Möller. A Survey on Music Listening and Management Behaviours. In Proc. ISMIR, pages 373--378, 2012.Google Scholar
- P. Lamere. I've got 10 million songs in my pocket: now what?. In Proc. RecSys 2012, pages 207--208, 20 Google ScholarDigital Library
- M. Kaminskas and F. Ricci. Contextual Music Information Retrieval and Recommendation: State ofthe Art and Challenges. Computer Science Review, 6(2--3):89--119, 2012.Google Scholar
- A. Lehtiniemi and J. Sepp\"anen. Evaluation of Automatic Mobile Playlist Generator. In Proc. MC, pages 452--459, 2007. Google ScholarDigital Library
- B. Logan. Content-Based Playlist Generation: Exploratory Experiments. In Proc. ISMIR, pages 295--296, 2002.Google Scholar
- B. McFee and G. R. G. Lanckriet. The Natural Language of Playlists. In Proc. ISMIR, pages 537--542, 2011.Google Scholar
- B. McFee and G. R. G. Lanckriet. Hypergraph Models of Playlist Dialects. In Proc. ISMIR, pages 343--348, 2012.Google Scholar
- J. Oh, S. Park, H. Yu, M. Song, and S. Park. Novel Recommendation Based on Personal Popularity Tendency. In Proc. ICDM '11, pages 507--516, 2011. Google ScholarDigital Library
- J. C. Platt, C. J. Burges, S. Swenson, C. Weare, and A. Zheng. Learning a Gaussian Process Prior for Automatically Generating MusicPlaylists. In Proc. NIPS, pages 1425--1432, 2001.Google Scholar
- G. Reynolds, D. Barry, T. Burke, and E. Coyle. Interacting With Large Music Collections: Towards the Use ofEnvironmental Metadata. In Proc. ICME, pages 989--992, 2008.Google ScholarCross Ref
- A. Said, B. J. Jain, and S. Albayrak. A 3D Approach to Recommender System Evaluation. In Proc. CSCW '13 Companion Volume, pages 263--266, 2013. Google ScholarDigital Library
- A. M. Sarroff and M. Casey. Modeling and Predicting Song Adjacencies In Commercial Albums. In Proc. SMC, 2012.Google Scholar
- M. Slaney and W. White. Measuring Playlist Diversity for Recommendation Systems. In Proc. AMCMM '06, pages 77--82, 2006. Google ScholarDigital Library
- L. Xiao, L. Lu, F. Seide, and J. Zhou. Learning a Music Similarity Measure on Automatic Annotations withApplication to Playlist Generation. In Proc. ICASSP, pages 1885--1888, 2009. Google ScholarDigital Library
Index Terms
- Biases in Automated Music Playlist Generation: A Comparison of Next-Track Recommending Techniques
Recommendations
Music Playlist Recommender System AFT-IS
ICCAE 2018: Proceedings of the 2018 10th International Conference on Computer and Automation EngineeringPreviously, we had proposed a playlist recommendation method to suggest music by considering the change in acoustic features in the song so that the transition between songs becomes smooth. Our previous method uses the last two songs in a playlist to ...
Listener-Inspired Automated Music Playlist Generation
RecSys '15: Proceedings of the 9th ACM Conference on Recommender SystemsThe objective of this PhD research is to deepen the understanding of how people listen to music and construct playlists. We believe that further insights into such mechanisms can lead to enhanced music recommendations. We research on the exploitation of ...
Music Playlist Recommendation Using Acoustic-Feature Transition Inside the Songs
MoMM2017: Proceedings of the 15th International Conference on Advances in Mobile Computing & MultimediaPreviously, we proposed a playlist recommendation method that recommends a music sequence that has smooth transitions of the acoustic features. Our previous method uses the last two songs in a playlist to make recommendations for following songs that ...
Comments