%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2016/08.12.00.56
%2 sid.inpe.br/sibgrapi/2016/08.12.00.56.22
%T Abordagem de Aprendizado Ativo para Classificação de Dados Biomédicos
%D 2016
%A Camargo, Guilherme,
%A Bugatti, Pedro Henrique,
%A Saito, Priscila Tiemi Maeda,
%@affiliation Universidade Tecnológica Federal do Paraná (UTFPR)
%@affiliation Universidade Tecnológica Federal do Paraná (UTFPR)
%@affiliation Universidade Tecnológica Federal do Paraná (UTFPR) e Universidade Estadual de Campinas (UNICAMP)
%E Aliaga, Daniel G.,
%E Davis, Larry S.,
%E Farias, Ricardo C.,
%E Fernandes, Leandro A. F.,
%E Gibson, Stuart J.,
%E Giraldi, Gilson A.,
%E Gois, João Paulo,
%E Maciel, Anderson,
%E Menotti, David,
%E Miranda, Paulo A. V.,
%E Musse, Soraia,
%E Namikawa, Laercio,
%E Pamplona, Mauricio,
%E Papa, João Paulo,
%E Santos, Jefersson dos,
%E Schwartz, William Robson,
%E Thomaz, Carlos E.,
%B Conference on Graphics, Patterns and Images, 29 (SIBGRAPI)
%C São José dos Campos, SP, Brazil
%8 4-7 Oct. 2016
%I Sociedade Brasileira de Computação
%J Porto Alegre
%S Proceedings
%K aprendizado ativo, análise de imagens, classificação, imagens biomédicas, floresta de caminhos ótimos.
%X A huge volume of biomedical data (images, genes, among others) is daily generated. The analysis of such data is a complex task that demands specialized knowledge, and the level of expertise directly impacts the diagnosis. Besides, due to the volume of data such task becomes extremely tiresome, and hence highly susceptible to errors. Trying to solve this problem, machine learning approaches have been proposed in the literature to perform automatic classification of such data. Despite the several proposed techniques, the great majority strictly focus just on the effectiveness, and relegate the efficiency of the classification. This paper presents a novel learning approach capable to obtain high accuracies, as well as maintaining a minimal involvement of the expert and interactive computational time during the learning process. To do so, the proposed approach exploits the active learning paradigm, in order to reduce, organize and select the most informative samples to the learning process of the pattern classifier.
%@language pt
%3 2016-sibgrapi-wip.pdf