BDBComp
Parceria:
SBC
Processo de seleção de atributos baseado em características métricas utilizando mineradores de dados e classificador neural

RAYMUNDO, E.M.Rodrigues, E.L.L.

This paper presents a research based characteristics metrics, using the database of images of carpal School of Engineering of São Carlos (EESC). The images were properly segmented to obtain the area, perimeter and length of each bone, thereby generating a database metric, the CarpEven. The information on the metric were applied to algorithms of data mining and subsequently subjected to neural classifiers, allowing a thorough research on attributes that are presented with potentially discriminatory behavior. This study attempts to find information that could be used in new methodologies for estimation of bone age.

http://www.lbd.dcc.ufmg.br/colecoes/wvc/2009/008.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