BDBComp
Parceria:
SBC
Splice Junction Recognition using Machine Learning Techniques

Ana C. LorenaGustavo E. A. P. A. BatistaAndré C. P. L. F. de CarvalhoMaria C. Monard

Since the start of the Human Genome Project, a large amount of sequence data has been generated. These data need to be analyzed. One of the main analysis to be carried out is the identification of regions of these sequences that correspond to genes. For such, one can search for particular signals associated with gene expression. Among the searched signals are the splice junctions. This recognition problem can be efficiently accomplished with the use of computational intelligent techniques. Many of the genetic databases, however, are characterized by the presence of high levels of noise, which can deteriorate the learning techniques' performance. This paper investigates the influence of noisy data in the performance of two different learning techniques (Decision Trees and Support Vector Machines), in the splice junction recognition problem. Results indicate that the elimination of noisy patterns from the datasets employed can improve Decision Trees' comprehensiveness and Support Vector Machines' performance.

http://www.lbd.dcc.ufmg.br:8080/colecoes/wob/2002/005.pdf

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD