New Iterative Approach for Solving Fully Fuzzy Polynomials

In this paper, a new architecture of fuzzy neural network (FNN) model is proposed in order to find a fuzzy solution of a fully fuzzy polynomial (FFP) with degree one. The proposed FNN is a two layer feed-forward neural network, that corresponds connection weight to output layer. The proposed architecture of artificial neural network can get a fuzzy input signal and calculates its corresponding fuzzy output. In order to find the approximate solution of these polynomials, first a cost function for the level sets of fuzzy output and fuzzy target is defined. Then for adjusting the three parameters of triangular fuzzy weight a learning algorithm which is based on the gradient descent method is introduced. The proposed method is illustrated by several examples with computer simulations. AMS subject classification: