BDBComp
Parceria:
SBC
CLASSIFICAÇÃO DE TOMATES UTILIZANDO REDES NEURAIS ARTIFICIAIS

B. DantasD. DéharbeS. GalvãoA. Martins MoreiraV. Medeiros Júnior

LOURO, Antônio Henrique FigueiraMENDONÇA, Michelle MagalhãesAdilson Gonzaga

This work describes a model-driven approach to design and develop software from the functional specification level down to assembly. The proposed approach builds upon the B method and provides a methodology to craft assembly-level software components in a rigorous way. While the B method is conventionally applied to produce algorithmic level software artifacts for safety-critical systems, it was not originally designed to handle the final transformations to source code and then to assembly. The users of the B method need thus to use code synthesis and compilation tools that do not offer the same rigorousness. Subtle bugs in these final steps may indeed jeopardize the whole engineering process. The approach proposed in the paper extends the B method to covers these last steps and therefore contributes to the scientific grand challenge of Computer Science proposed by Tony Hoare [6]: "The Verifying Compiler".

http://www.lbd.dcc.ufmg.br:8080/colecoes/semish/2008/014.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