Formulation of a multivalued recognition system

A recognition system based on fuzzy set theory and approximate reasoning has been described that is capable of handling various imprecise input patterns and providing a natural decision system. The input feature is considered to be of either quantative form or linguistic form or mixed form or set form. The entire feature space is decomposed here into some overlapping subdomains depending on the geometric structure and the relative position of the pattern classes found in the training samples. The various uncertainty (ambiguity) in the input statement has been managed by providinglmodifying membership values to a great extent. A relational matrix corresponding to the subdomains and the pattern classes has been considered in the modified Zadeh’s compositional rule of inference in order to recognize the samples. The linguistic output decision is associated with a confidence factor denoting the degree of certainty of a decision. The effectiveness of the algorithm has been demonstrated on some artificially generated patterns and also on the real life speech data. The recognition scores are described in terms of various choices namely, single correct, first correct, combined correct, second correct and fully wrong choices; thus provides a low rate of misclassification as compared to the conventional two-state systems.