Neuro-genetic approach to multidimensional fuzzy reasoning for pattern classification

To tackle the pattern classification problems first we give a new interpret ation to the multidimensional fuzzy implication (MFI). This new interpretat ion of MFI is used for multidimensional fuzzy reasoning (MFR) for pattern c lassification. We realize the new interpretation through multilayer percept ron. The learning scheme of the network is based on genetic algorithm (GA), A weight smoothing scheme is also proposed to improve neural network's gen eralization capability. The smoothing constraint is incorporated into the o bjective function of the network to reflect the neighborhood correlation an d to seek those solutions which have smooth connection weights. At the lear ning stage of the neural network fuzzy linguistic statements have been used . Once learned, the nonfuzzy features of a pattern can be classified using a fuzzy masking. The performance of the proposed scheme is tested through s ynthetic data. Finally, we apply the proposed scheme to the vowel recogniti on problem of one Indian language. (C) 2000 Elsevier Science B.V. All right s reserved.