Measuring the Performance of Evolutionary Multi-Objective Feature Selection for Prediction of Musical Genres and Styles

The prediction of high-level music categories, such as genres, styles, or personal preferences, helps to organise music collections. The relevance of single audio features for automatic classification depends on a certain category. Relevant feature subsets for each classification task can be identified by means of feature selection. Continuing our previous studies on multi-objective feature selection for music classification, in this work we measure an impact of evolutionary multi-objective feature selection on classification performance and compare it to the baseline application without feature selection. As confirmed by statistical tests, the integration of evolutionary multi-objective feature selection leads to a significant increase of performance according to both evaluation criteria as well as to classification error. This holds for all four tested classification methods and six music categories.

[1]  Günter Rudolph,et al.  Design and comparison of different evolution strategies for feature selection and consolidation in music classification , 2009, 2009 IEEE Congress on Evolutionary Computation.

[2]  Günter Rudolph,et al.  Multi-objective feature selection in music genre and style recognition tasks , 2011, GECCO '11.

[3]  อนิรุธ สืบสิงห์,et al.  Data Mining Practical Machine Learning Tools and Techniques , 2014 .

[4]  Katharina Morik,et al.  Automatic Feature Extraction for Classifying Audio Data , 2005, Machine Learning.

[5]  Mineichi Kudo,et al.  Comparison of algorithms that select features for pattern classifiers , 2000, Pattern Recognit..

[6]  C. Weihs,et al.  Statistical Comparison of Classifiers for Multi-objective Feature Selection in Instrument Recognition , 2012, Annual Conference of the Gesellschaft für Klassifikation.

[7]  Ichiro Fujinaga,et al.  Feature Selection Pitfalls and Music Classification , 2006, ISMIR.

[8]  Rainer Martin,et al.  Cepstral modulation ratio regression (CMRARE) parameters for audio signal analysis and classification , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Igor Vatolkin,et al.  AMUSE (Advanced MUSic Explorer) - A Multitool Framework for Music Data Analysis , 2010, ISMIR.

[10]  Tuomas Eerola,et al.  Generalizability and Simplicity as Criteria in Feature Selection: Application to Mood Classification in Music , 2011, IEEE Transactions on Audio, Speech, and Language Processing.

[11]  Kari Torkkola,et al.  Information-Theoretic Methods , 2006, Feature Extraction.

[12]  Meinard Müller,et al.  Towards Timbre-Invariant Audio Features for Harmony-Based Music , 2010, IEEE Transactions on Audio, Speech, and Language Processing.

[13]  Günter Rudolph,et al.  Multi-objective evolutionary feature selection for instrument recognition in polyphonic audio mixtures , 2012, Soft Comput..

[14]  Nicola Beume,et al.  An EMO Algorithm Using the Hypervolume Measure as Selection Criterion , 2005, EMO.

[15]  Grzegorz Rozenberg,et al.  Handbook of Natural Computing , 2011, Springer Berlin Heidelberg.

[16]  François Pachet,et al.  A taxonomy of musical genres , 2000, RIAO.

[17]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[18]  C. Emmanouilidis,et al.  A multiobjective evolutionary setting for feature selection and a commonality-based crossover operator , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[19]  Anil Kokaram,et al.  An Evaluation of Alternative Feature Selection Strategies and Ensemble Techniques for Classifying Music , 2003 .

[20]  Simon Dixon,et al.  Approximate Note Transcription for the Improved Identification of Difficult Chords , 2010, ISMIR.

[21]  Mike Preuss,et al.  Operator Learning for a Problem Class in a Distributed Peer-to-Peer Environment , 2002, PPSN.

[22]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[23]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems , 2002, Genetic Algorithms and Evolutionary Computation.

[24]  Igor Vatolkin,et al.  Partition Based Feature Processing for Improved Music Classification , 2010, GfKl.

[25]  Ichiro Fujinaga,et al.  Machine recognition of timbre using steady-state tone of acoustic musical instruments , 1998, ICMC.

[26]  Philip H. Ramsey.  Nonparametric Statistical Methods , 1974, Technometrics.