Este post comenta a aula 3 sobre Direitos Humanos da Universidade da Califórnia em Berkley. O curso inteiro está disponível na Internet. Link aqui. As aulas foram dadas pelo professor Thomas W. Laqueur.
Nesta aula, o convidado especial é o prof. Kinch Hoekstra. E o tópico é direitos humanos antes de "Direitos Humanos". Isto é, até o século XVIII.
Hoekstra comenta como era comum a ideia de direito do vencedor, onde depois de uma guerra o vencedor tinha direitos morais e absolutos sobre os vencidos para fazer o que quisessem com eles. Um pensamento assustador. Em quais povos e em quais momentos na História moderna poderíamos identificar este sentimento? Eu diria no Colonialismo europeu, por exemplo. A ideia dos Estados Unidos de direito sobre Cuba nos séculos XIX e XX (mesmo sem ganhar a guerra, os EUA sentiram ter direito sobre a ilha.)
Mais adiante, Hoekstra e Laqueur comentam sobre Cristandade, que vem justamente atrelada a ideia de domínio. Isto é, direito divino sobre o outro.
Outra ideia claramente explicada por Hekstra é a da justificativa de "inferioridade moral com base natural" de outro povo usada para subjugá-lo. Laqueur muito oportunamente também comentou como o próprio discurso de "humanidade inata" ou "humanidade geral" pode ser adulterado para justificar crimes contra um povo, ou minoria. Por exemplo: os índios da América têm alma (caso específico: os índios da América na discussão Bartolomé de las Casas versus Sepúlveda) (esqueci a bula Papal). Já que eles têm alma, então temos o dever moral de convertê-los e salvá-los do Inferno.
Não precisamos ir muito longe. Note como o vídeo do pastor Paschoal Piragine Jr retrata índios brasileiros. Note também como os gays são caricaturados por evangélicos e católicos.
Showing posts with label academic. Show all posts
Showing posts with label academic. Show all posts
Tuesday, March 22, 2011
Friday, June 15, 2007
My Masters Thesis
For those that are looking for something to read, here's my Master's Thesis.
FEATURE SELECTION AND EVALUATION FOR GENRE CLASSIFICATION OF
SYMBOLICALLY ENCODED CLASSICAL MUSIC WITH THE AID OF MACHINE
LEARNING
Gustavo Cesar de Souza Frederico
Thesis submitted to the Faculty of Graduate and Postdoctoral Studies
in partial fulfillment of the requirements for the degree of Master of Computer Science
June 1st, 2006
Ottawa-Carleton Institute for Computer Science
School of Information Technology and Engineering
University of Ottawa
Abstract
This work defines useful features for the classification of symbolically encoded music into 14 classical genres namely chorale, symphony, étude, fugue, prelude, contrafactum, sonata, mazurka, motet, sonatina, waltze, concerto, Gregorian chant and scherzo. Features are based on Music Theory and grouped into seven categories: distances in the harmonic möbius strip, distances on the line of fifths, scale, rhythmic syncopation and meter, polyphony measurements, duration and instrumentation. Features are extracted and ranked combining 5 filter-based methods. Six Machine Learning algorithms are defined for classification: three Support Vector Machines, one Bayesian network, the C4.5 and random forests. Using nested cross-validation for training and testing and considering all the features, the Bayesian network classifier yields 84.10 % empirical accuracy. The FEATUROMETRE process measures the usefulness of the feature subsets in an approach similar to wrapper methods, conveying relevant information to domain experts. Another experiment measures the usefulness and accuracy of features individually and by category using FEATUROMETRE. Grouping the music pieces by their period, the measured accuracy with the random forest classifier in the second experiment reaches 89.81 %.
FEATURE SELECTION AND EVALUATION FOR GENRE CLASSIFICATION OF
SYMBOLICALLY ENCODED CLASSICAL MUSIC WITH THE AID OF MACHINE
LEARNING
Gustavo Cesar de Souza Frederico
Thesis submitted to the Faculty of Graduate and Postdoctoral Studies
in partial fulfillment of the requirements for the degree of Master of Computer Science
June 1st, 2006
Ottawa-Carleton Institute for Computer Science
School of Information Technology and Engineering
University of Ottawa
Abstract
This work defines useful features for the classification of symbolically encoded music into 14 classical genres namely chorale, symphony, étude, fugue, prelude, contrafactum, sonata, mazurka, motet, sonatina, waltze, concerto, Gregorian chant and scherzo. Features are based on Music Theory and grouped into seven categories: distances in the harmonic möbius strip, distances on the line of fifths, scale, rhythmic syncopation and meter, polyphony measurements, duration and instrumentation. Features are extracted and ranked combining 5 filter-based methods. Six Machine Learning algorithms are defined for classification: three Support Vector Machines, one Bayesian network, the C4.5 and random forests. Using nested cross-validation for training and testing and considering all the features, the Bayesian network classifier yields 84.10 % empirical accuracy. The FEATUROMETRE process measures the usefulness of the feature subsets in an approach similar to wrapper methods, conveying relevant information to domain experts. Another experiment measures the usefulness and accuracy of features individually and by category using FEATUROMETRE. Grouping the music pieces by their period, the measured accuracy with the random forest classifier in the second experiment reaches 89.81 %.
Subscribe to:
Posts (Atom)