LNAI 2903 MML Classification of Music Genres 1st
Edition by Adrian Bickerstaffe, Enes Makalic
ISBN 9783540206460 354020646X pdf download
https://ebookball.com/product/lnai-2903-mml-classification-of-
music-genres-1st-edition-by-adrian-bickerstaffe-enes-makalic-
isbn-9783540206460-354020646x-14280/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
LNAI 2903 Pareto Neuro Ensembles 1st Edition by Hussein Abbass ISBN
9783540206460 354020646X
https://ebookball.com/product/lnai-2903-pareto-neuro-
ensembles-1st-edition-by-hussein-abbass-
isbn-9783540206460-354020646x-9200/
LNAI 2903 Model Based Reinforcement Learning for Alternating Markov
Games 1st Edition by Drew Mellor ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-model-based-
reinforcement-learning-for-alternating-markov-games-1st-edition-
by-drew-mellor-isbn-9783540206460-354020646x-10942/
LNAI 2903 Race Car Chassis Tuning Using Artificial Neural Networks 1st
Edition by David Butler, Vishy Karri ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-race-car-chassis-tuning-
using-artificial-neural-networks-1st-edition-by-david-butler-
vishy-karri-isbn-9783540206460-354020646x-11378/
LNAI 2903 Dynamic Variable Filtering for Hard Random 3SAT Problems 1st
Edition by Anbulagan, John Thornton, Abdul Sattar ISBN 9783540206460
354020646X
https://ebookball.com/product/lnai-2903-dynamic-variable-
filtering-for-hard-random-3sat-problems-1st-edition-by-anbulagan-
john-thornton-abdul-sattar-isbn-9783540206460-354020646x-14526/
,LNAI 2903 Translating Novelty of Business Model into Terms of Modal
Logics 1st Edition by Hiroshi Kawakami, Ryosuke Akinaga, Hidetsugu
Suto, Osamu Katai ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-translating-novelty-of-
business-model-into-terms-of-modal-logics-1st-edition-by-hiroshi-
kawakami-ryosuke-akinaga-hidetsugu-suto-osamu-katai-
isbn-9783540206460-354020646x-9332/
LNAI 2903 Towards Automated Creation of Image Interpretation Systems
1st Edition by Ilya Levner, Vadim Bulitko, Lihong Li, Greg Lee,
Russell Greiner ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-towards-automated-
creation-of-image-interpretation-systems-1st-edition-by-ilya-
levner-vadim-bulitko-lihong-li-greg-lee-russell-greiner-
isbn-9783540206460-354020646x-9084/
LNAI 2903 Robustness for Evaluating Rule Generalization Capability in
Data Mining 1st Edition by Dianhui Wang, Tharam Dillon, Xiaohang Ma
ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-robustness-for-
evaluating-rule-generalization-capability-in-data-mining-1st-
edition-by-dianhui-wang-tharam-dillon-xiaohang-ma-
isbn-9783540206460-354020646x-11134/
LNAI 2903 Fuzzy Cognitive Map Learning Based on Nonlinear Hebbian Rule
1st Edition by Elpiniki Papageorgiou, Chrysostomos Stylios, Peter
Groumpos ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-fuzzy-cognitive-map-
learning-based-on-nonlinear-hebbian-rule-1st-edition-by-elpiniki-
papageorgiou-chrysostomos-stylios-peter-groumpos-
isbn-9783540206460-354020646x-10728/
LNAI 2903 Reduction of Non Deterministic Automata for Hidden Markov
Model Based Pattern Recognition Applications 1st Edition by Frederic
Maire, Frank Wathne, Alain Lifchitz ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-reduction-of-non-
deterministic-automata-for-hidden-markov-model-based-pattern-
recognition-applications-1st-edition-by-frederic-maire-frank-
wathne-alain-lifchitz-isbn-9783540206460-354020646x/
, MML Classification of Music Genres
Adrian C. Bickerstaffe and Enes Makalic
School of Computer Science and Software Engineering
Monash University (Clayton Campus)
Clayton, Victoria 3800, Australia
Abstract. Inference of musical genre, whilst seemingly innate to the hu-
man mind, remains a challenging task for the machine learning commu-
nity. Online music retrieval and automatic music generation are just two
of many interesting applications that could benefit from such research.
This paper applies four different classification methods to the task of dis-
tinguishing between rock and classical music styles. Each method uses the
Minimum Message Length (MML) principle of statistical inference. The
first, an unsupervised learning tool called Snob, performed very poorly.
Three supervised classification methods, namely decision trees, decision
graphs and neural networks, performed significantly better. The defining
attributes of the two musical genres were found to be pitch mean and
standard deviation, duration mean and standard deviation, along with
counts of distinct pitches and rhythms per piece. Future work includes
testing more attributes for significance, extending the classification to
include more genres (for example, jazz, blues etcetera) and using proba-
bilistic (rather than absolute) genre class assignment. Our research shows
that the distribution of note pitch and duration can indeed distinguish
between significantly different types of music.
1 Introduction
The task of successfully identifying music genres, while trivial for humans, is
difficult to achieve using machine learning techniques. However, applications of
automated music genre recognition are numerous and significant. For example,
a large database of music from unknown sources could be arranged to facilitate
fast searching and retrieval. To illustrate, retrieval of different pieces from the
same genre would become easily possible. Successful models of musical genres
would also be of great interest to musicologists. Discovering the attributes that
define a genre would provide insight to musicians and assist in automatically
generating pieces of a particular style.
Research toward music classification is reasonably well-established. Soltau
et al. developed a music style classifier using a three-layer feedforward neural
network and temporal modelling [1]. The classifier was trained using raw audio
samples from four genres of music: rock, pop, techno and classical. Cilibrasi et
Author list order determined stochastically.
T.D. Gedeon and L.C.C. Fung (Eds.): AI 2003, LNAI 2903, pp. 1063–1071, 2003.
c Springer-Verlag Berlin Heidelberg 2003
Edition by Adrian Bickerstaffe, Enes Makalic
ISBN 9783540206460 354020646X pdf download
https://ebookball.com/product/lnai-2903-mml-classification-of-
music-genres-1st-edition-by-adrian-bickerstaffe-enes-makalic-
isbn-9783540206460-354020646x-14280/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
LNAI 2903 Pareto Neuro Ensembles 1st Edition by Hussein Abbass ISBN
9783540206460 354020646X
https://ebookball.com/product/lnai-2903-pareto-neuro-
ensembles-1st-edition-by-hussein-abbass-
isbn-9783540206460-354020646x-9200/
LNAI 2903 Model Based Reinforcement Learning for Alternating Markov
Games 1st Edition by Drew Mellor ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-model-based-
reinforcement-learning-for-alternating-markov-games-1st-edition-
by-drew-mellor-isbn-9783540206460-354020646x-10942/
LNAI 2903 Race Car Chassis Tuning Using Artificial Neural Networks 1st
Edition by David Butler, Vishy Karri ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-race-car-chassis-tuning-
using-artificial-neural-networks-1st-edition-by-david-butler-
vishy-karri-isbn-9783540206460-354020646x-11378/
LNAI 2903 Dynamic Variable Filtering for Hard Random 3SAT Problems 1st
Edition by Anbulagan, John Thornton, Abdul Sattar ISBN 9783540206460
354020646X
https://ebookball.com/product/lnai-2903-dynamic-variable-
filtering-for-hard-random-3sat-problems-1st-edition-by-anbulagan-
john-thornton-abdul-sattar-isbn-9783540206460-354020646x-14526/
,LNAI 2903 Translating Novelty of Business Model into Terms of Modal
Logics 1st Edition by Hiroshi Kawakami, Ryosuke Akinaga, Hidetsugu
Suto, Osamu Katai ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-translating-novelty-of-
business-model-into-terms-of-modal-logics-1st-edition-by-hiroshi-
kawakami-ryosuke-akinaga-hidetsugu-suto-osamu-katai-
isbn-9783540206460-354020646x-9332/
LNAI 2903 Towards Automated Creation of Image Interpretation Systems
1st Edition by Ilya Levner, Vadim Bulitko, Lihong Li, Greg Lee,
Russell Greiner ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-towards-automated-
creation-of-image-interpretation-systems-1st-edition-by-ilya-
levner-vadim-bulitko-lihong-li-greg-lee-russell-greiner-
isbn-9783540206460-354020646x-9084/
LNAI 2903 Robustness for Evaluating Rule Generalization Capability in
Data Mining 1st Edition by Dianhui Wang, Tharam Dillon, Xiaohang Ma
ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-robustness-for-
evaluating-rule-generalization-capability-in-data-mining-1st-
edition-by-dianhui-wang-tharam-dillon-xiaohang-ma-
isbn-9783540206460-354020646x-11134/
LNAI 2903 Fuzzy Cognitive Map Learning Based on Nonlinear Hebbian Rule
1st Edition by Elpiniki Papageorgiou, Chrysostomos Stylios, Peter
Groumpos ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-fuzzy-cognitive-map-
learning-based-on-nonlinear-hebbian-rule-1st-edition-by-elpiniki-
papageorgiou-chrysostomos-stylios-peter-groumpos-
isbn-9783540206460-354020646x-10728/
LNAI 2903 Reduction of Non Deterministic Automata for Hidden Markov
Model Based Pattern Recognition Applications 1st Edition by Frederic
Maire, Frank Wathne, Alain Lifchitz ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-reduction-of-non-
deterministic-automata-for-hidden-markov-model-based-pattern-
recognition-applications-1st-edition-by-frederic-maire-frank-
wathne-alain-lifchitz-isbn-9783540206460-354020646x/
, MML Classification of Music Genres
Adrian C. Bickerstaffe and Enes Makalic
School of Computer Science and Software Engineering
Monash University (Clayton Campus)
Clayton, Victoria 3800, Australia
Abstract. Inference of musical genre, whilst seemingly innate to the hu-
man mind, remains a challenging task for the machine learning commu-
nity. Online music retrieval and automatic music generation are just two
of many interesting applications that could benefit from such research.
This paper applies four different classification methods to the task of dis-
tinguishing between rock and classical music styles. Each method uses the
Minimum Message Length (MML) principle of statistical inference. The
first, an unsupervised learning tool called Snob, performed very poorly.
Three supervised classification methods, namely decision trees, decision
graphs and neural networks, performed significantly better. The defining
attributes of the two musical genres were found to be pitch mean and
standard deviation, duration mean and standard deviation, along with
counts of distinct pitches and rhythms per piece. Future work includes
testing more attributes for significance, extending the classification to
include more genres (for example, jazz, blues etcetera) and using proba-
bilistic (rather than absolute) genre class assignment. Our research shows
that the distribution of note pitch and duration can indeed distinguish
between significantly different types of music.
1 Introduction
The task of successfully identifying music genres, while trivial for humans, is
difficult to achieve using machine learning techniques. However, applications of
automated music genre recognition are numerous and significant. For example,
a large database of music from unknown sources could be arranged to facilitate
fast searching and retrieval. To illustrate, retrieval of different pieces from the
same genre would become easily possible. Successful models of musical genres
would also be of great interest to musicologists. Discovering the attributes that
define a genre would provide insight to musicians and assist in automatically
generating pieces of a particular style.
Research toward music classification is reasonably well-established. Soltau
et al. developed a music style classifier using a three-layer feedforward neural
network and temporal modelling [1]. The classifier was trained using raw audio
samples from four genres of music: rock, pop, techno and classical. Cilibrasi et
Author list order determined stochastically.
T.D. Gedeon and L.C.C. Fung (Eds.): AI 2003, LNAI 2903, pp. 1063–1071, 2003.
c Springer-Verlag Berlin Heidelberg 2003
