Abstract
The traditional role of nearest-neighbor classification in music classification research is that of a straw man opponent for the learning approach of the hour. Recent work in high-dimensional indexing has shown that approximate nearest-neighbor algorithms are extremely scalable, yielding results of reasonable quality from billions of high-dimensional features. With such efficient large-scale classifiers, the traditional music classification methodology of aggregating and compressing the audio features is incorrect; instead the approximate nearest-neighbor classifier should be given an extensive data collection to work with. We present a case study, using a well-known MIR classification benchmark with well-known music features, which shows that a simple nearest-neighbor classifier performs very competitively when given ample data. In this position paper, we therefore argue that nearest-neighbor classification has been treated unfairly in the literature and may be much more competitive than previously thought.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aucouturier, J.J., Pachet, F.: Representing musical genre: a state of the art. J. New Music Res. 32(1), 83–93 (2003)
Babenko, A., Lempitsky, V.S.: The inverted multi-index. TPAMI 37(6), 1247–1260 (2015)
Babenko, A., Lempitsky, V.S.: Efficient indexing of billion-scale datasets of deep descriptors. In: Proceedings of CVPR, Las Vegas, NV, USA (2016)
Bergstra, J., Casagrande, N., Erhan, D., Eck, D., Kégl, B.: Aggregate features and ADABOOST for music classification. Mach. Learn. 65(2–3), 473–484 (2006)
Boiman, O., Shechtman, E., Irani, M.: In defense of nearest-neighbor based image classification. In: Proceedings of CVPR (2008)
Fabbri, F.: A theory of musical genres: two applications. Popular Music Perspect. 1, 52–81 (1981)
Gu\(\eth \)mundsson, G., Amsaleg, L., Jónsson, B.: Distributed high-dimensional index creation using Hadoop, HDFS and C++. In: Proceedings of CBMI (2012)
Gu\(\eth \)mundsson, G., Amsaleg, L., Jónsson, B., Franklin, M.J.: Towards engineering a web-scale multimedia service: a case study using Spark. In: Proceedings of MMSys (2017)
Jégou, H., Douze, M., Schmid, C.: Product quantization for nearest neighbor search. TPAMI 33(1), 117–128 (2011)
Knees, P., Schedl, M.: A survey of music similarity and recommendation from music context data. ACM TOMCCAP 10(1), 1–21 (2013)
Knees, P., Schedl, M.: Music Similarity and Retrieval - An Introduction to Audio- and Web-based Strategies. Springer, Heidelberg (2016)
Lejsek, H., Jónsson, B., Amsaleg, L.: NV-Tree: nearest neighbours at the billion scale. In: Proceedings of ICMR (2011)
Li, T., Ogihara, M., Li, Q.: A comparative study on content-based music genre classification. In: Proceedings of SIGIR, Toronto, Canada (2003)
Moise, D., Shestakov, D., Gu\(\eth \)mundsson, G., Amsaleg, L.: Indexing and searching 100M images with map-reduce. In: Proceedings of ICMR (2013)
van den Oord, A., Dieleman, S., Schrauwen, B.: Deep content-based music recommendation. In: Proceedings of NIPS (2013)
van den Oord, A., Dieleman, S., Schrauwen, B.: Transfer learning by supervised pre-training for audio-based music classification. In: Proceedings of ISMIR (2014)
Pachet, F., Cazaly, D.: A taxonomy of musical genre. In: Proceedings of RIAO (2000)
Pálmason, H., Jónsson, B., Schedl, M., Knees, P.: Music genre classification revisited: An in-depth examination guided by music experts. In: Proceedings of CMMR (2017)
Panagakis, I., Benetos, E., Kotropoulos, C.: Music genre classification: a multilinear approach. In: Proceedings of ISMIR (2008)
Panagakis, Y., Kotropoulos, C., Arce, G.R.: Music genre classification via sparse representations of auditory temporal modulations. In: Proceedings of EUSIPCO (2009)
Prockup, M., Ehmann, A.F., Gouyon, F., Schmidt, E.M., Celma, Ò., Kim, Y.E.: Modeling genre with the music genome project: comparing human-labeled attributes and audio features. In: Proceedings of ISMIR (2015)
Schnitzer, D.: Indexing content-based music similarity models for fast retrieval in massive databases. Dissertation, Johannes Kepler University, Austria (2012)
Seyerlehner, K., Schedl, M., Knees, P., Sonnleitner, R.: A refined block-level feature set for classification, similarity and tag prediction. In: Proceedings of MIREX (2011)
Seyerlehner, K., Widmer, G., Pohle, T.: Fusing block-level features for music similarity estimation. In: Proceedings of Digital Audio Effects (DAFx) (2010)
Sturm, B.L.: An analysis of the GTZAN music genre dataset. In: Proceedings of MIRUM (2012)
Sturm, B.L.: Classification accuracy is not enough. JIIS 41, 371–406 (2013)
Sturm, B.L.: A survey of evaluation in music genre recognition. In: Nürnberger, A., Stober, S., Larsen, B., Detyniecki, M. (eds.) AMR 2012. LNCS, vol. 8382, pp. 29–66. Springer, Cham (2014). doi:10.1007/978-3-319-12093-5_2
Tzanetakis, G., Cook, P.: Musical genre classification of audio signals. IEEE Trans. Speech Audio Process. 10(5), 293–302 (2002)
Acknowledgements
This work was partially supported by Icelandic Student Research Fund grant 100390001, Austrian Science Fund (FWF) grant P25655 and Austrian FFG grant 858514 (SmarterJam).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Pálmason, H., Jónsson, B.Þ., Amsaleg, L., Schedl, M., Knees, P. (2017). On Competitiveness of Nearest-Neighbor-Based Music Classification: A Methodological Critique. In: Beecks, C., Borutta, F., Kröger, P., Seidl, T. (eds) Similarity Search and Applications. SISAP 2017. Lecture Notes in Computer Science(), vol 10609. Springer, Cham. https://doi.org/10.1007/978-3-319-68474-1_19
Download citation
DOI: https://doi.org/10.1007/978-3-319-68474-1_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-68473-4
Online ISBN: 978-3-319-68474-1
eBook Packages: Computer ScienceComputer Science (R0)