Interpretability of Music Classification as a Criterion for Evolutionary Multi-objective Feature Selection

详细信息    查看全文

• 作者：Igor Vatolkin (16)
  G眉nter Rudolph (16)
  Claus Weihs (17)
  
  16. Department of Computer Science ; TU Dortmund ; Dortmund ; Germany
  17. Faculty of Statistics ; TU Dortmund ; Dortmund ; Germany
  
• 关键词：Interpretable music classification ; Evolutionary multi ; objective optimisation ; Feature selection
• 刊名：Lecture Notes in Computer Science
• 出版年：2015
• 出版时间：2015
• 年：2015
• 卷：9027
• 期：1
• 页码：236-248
• 全文大小：1,335 KB
• 参考文献：1. Anglade, A, Benetos, E, Mauch, M, Dixon, S (2010) Improving music genre classification using automatically induced harmony rules. J. New Music Res. 39: pp. 349-361 CrossRef
  2. Arabi A.F., Lu, G.: Enhanced polyphonic music genre classification using high level features. In: Proceedings of the IEEE International Conference on Signal and Image Processing Applications (ICSIPA), pp. 101鈥?06 (2009)
  3. Van Balen, J., Bountouridis, D., Wiering, F., Veltkamp, R.: Cognition-inspired descriptors for scalable cover song retrieval. In: Proceedings of 15th International Society for Music Information Retrieval Conference, pp. 379鈥?84 (2014)
  4. Berenzweig, A., Ellis, D.P.W., Lawrence, S.: Anchor space for classification and similarity measurement of music. In: Proceedings of the IEEE International Conference on Multimedia and Expo, pp. 29鈥?2 (2003)
  5. Beume, N, Naujoks, B, Emmerich, M (2007) SMS-EMOA: multiobjective selection based on dominated hypervolume. Eur. J. Sci. Res. 181: pp. 1653-1669
  6. Bischl, B, Vatolkin, I, Preuss, M Selecting small audio feature sets in music classification by means of asymmetric mutation. In: Schaefer, R, Cotta, C, Ko艂odziej, J, Rudolph, G eds. (2010) Parallel Problem Solving from Nature, PPSN XI. Springer, Heidelberg, pp. 314-323
  7. Deb, K (2001) Multi-objective Optimization Using Evolutionary Algorithms. Wiley, Chichester
  8. Ellis, K, Coviello, E, Chan, A, Lanckriet, G (2013) A bag of systems representation for music auto-tagging. IEEE Trans. Audio Speech Lang. Process. 21: pp. 2554-2569 CrossRef
  9. Fern谩ndez, F, Ch谩vez, F Fuzzy rule based system ensemble for music genre classification. In: Machado, P, Romero, J, Carballal, A eds. (2012) Evolutionary and Biologically Inspired Music, Sound, Art and Design. Springer, Heidelberg, pp. 84-95 CrossRef
  10. Herlands, W., Der, R., Greenberg, Y., Levin, S.: A machine learning approach to musically meaningful homogeneous style classification. In: Proceedings of the 28th AAAI Conference on Artificial Intelligence (AAAI), pp. 276鈥?82 (2014)
  11. Kostek, B, Plewa, M (2013) Parametrisation and correlation analysis applied to music mood classification. Int. J. Comput. Intell. Stud. 2: pp. 4-25 CrossRef
  12. Kudo, M, Sklansky, J (2000) Comparison of algorithms that select features for pattern classifiers. Pattern Recogn. 33: pp. 25-41 CrossRef
  13. Lidy, L., Rauber, A., Pertusa, A., I帽esta, J.M.: Improving genre classification by combination of audio and symbolic descriptors using a transcription system. In: Proceedings of the 8th International Conference on Music Information Retrieval (ISMIR), pp. 61鈥?6 (2007)
  14. Mauch, M., Levy, M.: Structural change on multiple time scales as a correlate of musical complexity. In: Proceedings 12th International Society for Music Information Retrieval Conference (ISMIR), pp. 489鈥?94 (2011)
  15. McKay, C.: Automatic Music Classification with jMIR. Ph.D. thesis, Department of Music Research, Schulich School of Music, McGill University (2010)
  16. McKay, C., Burgoyne, J.A., Hockman, J., Smith, J.B.L., Vigliensoni, G., Fujinaga, I.: Evaluating the genre classification performance of lyrical features relative to audio, symbolic and cultural features. In: Proceedings of the 11th International Society for Music Information Retrieval Conference (ISMIR), pp. 213鈥?18 (2010)
  17. Mierswa, I, Morik, K (2005) Automatic feature extraction for classifying audio data. Mach. Learn. J. 58: pp. 127-149 CrossRef
  18. Miranda, ER, Biles, JA (2007) Evolutionary Computer Music. Springer, London CrossRef
  19. Mukhopadhyay, A, Maulik, U, Bandyopadhyay, S, Coello Coello, CA (2014) A survey of multiobjective evolutionary algorithms for data mining: Part I. IEEE Trans. Evol. Comput. 18: pp. 4-19 CrossRef
  20. Pachet, F, Zils, A Evolving automatically high-level music descriptors from acoustic signals. In: Wiil, UK eds. (2004) Computer Music Modeling and Retrieval. Springer, Heidelberg, pp. 42-53 CrossRef
  21. Panda, R., Rocha, B., Paiva, R.P.: Dimensional music emotion recognition: combining standard and melodic audio features. In: Proceedings of the 10th International Symposium on Computer Music Multidisciplinary Research (CMMR), pp. 583鈥?93 (2013)
  22. R枚tter, G, Vatolkin, I, Weihs, C Computational prediction of high-level descriptors of music personal categories. In: Lausen, B, Poel, D, Ultsch, A eds. (2013) Algorithms from and for Nature and Life. Springer, Heidelberg, pp. 529-537 CrossRef
  23. Vatolkin, I.: Improving Supervised Music Classification by Means of Multi-Objective Evolutionary Feature Selection. Ph.D. thesis, TU Dortmund (2013)
  24. Zanoni, M., Ciminieri, D., Sarti, A. Tubaro, S.: Searching for dominant high-level features for music information retrieval. In: Proceedings of the 20th European Signal Processing Conference (EUSIPCO), pp. 2025鈥?029 (2012)
  25. Zhu, J., Xue, X., Lu, H.: Musical genre classification by instrumental features. In: Proceedings of the 2004 International Computer Music Conference (ICMC), pp. 580鈥?83 (2004)
  26. Zitzler, E, Thiele, L Multiobjective optimization using evolutionary algorithms - a comparative case study. In: Eiben, AE, B盲ck, T, Schoenauer, M, Schwefel, H-P eds. (1998) Parallel Problem Solving from Nature - PPSN V. Springer, Heidelberg, pp. 292-304 CrossRef
  
• 作者单位：Evolutionary and Biologically Inspired Music, Sound, Art and Design
• 丛书名：978-3-319-16497-7
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349

文摘

The development of numerous audio signal characteristics led to an increase of classification performance for automatic categorisation of music audio recordings. Unfortunately, models built with such low-level descriptors lack of interpretability. Musicologists and listeners can not learn musically meaningful properties of genres, styles, composers, or personal preferences. On the other side, there are new algorithms for the mining of interpretable features from music data: instruments, moods and melodic properties, tags and meta data from the social web, etc. In this paper, we propose an approach how evolutionary multi-objective feature selection can be applied for a systematic maximisation of interpretability without a limitation to the usage of only interpretable features. We introduce a simple hypervolume based measure for the evaluation of trade-off between classification performance and interpretability and discuss how the results of our study may help to search for particularly relevant high-level descriptors in future.