An interval fuzzy model using a fuzzy neural network

The authors propose a model of human decision making, based on fuzzy inference whose consequences are described with weighted linear equations. In this model, the input space is divided into fuzzy sub-spaces, and the input-output relationships are identified with a weighted linear equation in each sub-space. The new fuzzy neural network (FNN) identifies the upper side and lower side of the interval fuzzy model. From the identified model, it is easy to determine which attributes mainly give positive/negative evaluations of the objects. Experiments were conducted using face graphs to show the feasibility of the new model and the new FNN