Bruno P. de Amorim,  Cleber Zanchettin,  Denise M. R. H. Vasconcellos,  Germano C. Vasconcelos,  Teresa B. Ludermir,  Aluízio F. R. Araújo. 

This work presents an evaluation of the Feature-Weighted Detector (FWD) network, a neuro-fuzzy model for pattern classification, relevant feature selection and extraction of fuzzy If-Then rules. An experimental investigation was conducted with the FWD model, employing two medical databases broadly used in the validation of knowledge extraction models. The results achieved showed the effectiveness of this model when compared to Multi-Layer Perceptron (MLP) networks in relation to pattern classification, besides demonstrating its ability to feature selection and rule extraction, unavailable characteristics in MLP networks. Clique no link abaixo para buscar o texto completo deste trabalho na Web: Buscar na Web Processando consulta.