How to Cite
Abstract
Este trabalho propõe uma nova técnica para a maximização de margem, através do tratamento de dados. As várias abordagens de tratamento de dados serão utilizadas no treinamento do Perceptron Espanhol. As distâncias entre os vetores de suporte e a superfície de separação foram utilizadas como métrica para a avaliação da máxima margem. Os vetores de suporte são fornecidos através da Support Vector Machine (SVM), após treinamento com a base de dados sem tratamento. A base utilizada é sintética, linearmente separável, com duas classes, aleatória de distribuição normal. Os resultados sugerem que a nova técnica proporciona uma melhoria na margem gerada com o perceptron espanhol, quando o erro de treinamento é maior.
Palavras-chave: Maximização de margens. Perceptron Espanhol. SVM.
Neural networks in the data processing for maximizing
Abstract
This paper proposes a new technique for maximizing margin, through the processing of data. The various data processing approaches will be used in Spanish Perceptron training. The distances between the support vectors and the separation surface was used as a metric to evaluate the maximum margin. The support vectors are provided by the Support Vector Machine (SVM) after training database without treatment. The base used is synthetic linearly separable, with two classes, random normal distribution. The results suggest that the new technique provides an improvement in margin generated with Spanish Perceptron, when the training error is higher.
Keywords: Maximizing margins. Spanish Perceptron. SVM.
References
BREIMAN, L. Bagging predictors. Technical Report, Berkeley, n. 421, Sept. 1994.
FERNANDEZ-DELGADO, M. et al. Direct parallel perceptrons (DPPs): fast analytical calculation of the parallel perceptrons weights with margin control for classification tasks. IEEE transactions on neural networks, v. 22, n. 11, p. 1837-1838, Nov. 2011.
HORTA, E. G. Aplicação de máquinas de aprendizado extremo ao problema de aprendizado ativo. Belo Horizonte: Centro de Desenvolvimento e Planejamento Regional, Universidade Federal de Minas Gerais, 2013.
KARATZOGLOU, A.; SMOLA, A.; HORNIK, Kurt. Kernlab: an S4 Package for Kernel Methods in R. Journal of Statistical Software. v. 11, Apr. 2004.
MONTGOMERY, D. C.; RUNGER, G. C. Estatística aplicada e probabilidade para engenheiros. 8. ed. Rio de Janeiro: LTC, 2008.
SMOLA, A. et al. Advances in Large Margin Classifiers. Cambridge, Massachusetts: MIT, 2000. 422 p.
VAPNIK V; CORTES C. Support-Vector Networks. Machine Learning, v. 20, n. 3, p. 273-297, Sep. 1995.
Authors who post on ForScience agree to the following terms:
- The said work is of own authorship and has never been published or sent to another magazine, this magazine having the right of exclusivity on the edition and publication, under the license Creative Commons Attribution License 3.0 (CC-BY-NC) which allows anyone to share to adapt the work according to the following terms:
Atribuição — You must give the appropriate credit, provide a link to the license and indicate if the changes have been made. You must do so in any reasonable circumstance, but in no way do you suggest to the licensor to support you or your use.
Non-Commercial — You may not use the material for commercial purposes.
- If plagiarism has been proven in any published work, this Journal is free of any responsibility, and its author (s) shall bear the penalties provided by law.
- Authors are authorized to take additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (eg, publish in institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
- Authors are allowed and encouraged to publish and distribute their work online (eg in institutional repositories or on their personal page), as this can generate productive changes, as well as increase the impact and citation of the published work. However, it should be noted that once approved by the evaluators, the manuscript can not be changed. If the author wishes to do so, he should restart the submission process.