Neuro-fuzzy network based on rough sets and its applications

A new constructive method of the neuro-fuzzy network based on rough sets is proposed. First, an initial fuzzy rule base is generated from the history input-output data pairs by rough sets approach. Then, a neuro-fuzzy network is formed according to the rule table. And the learning algorithm based on the gradient descent method is given. The major advantage of this approach is to optimize the overall structure of the neuro-fuzzy network as well as to adjust each parameter of fuzzy rules without doing the complicated clustering process. Finally, the efficiency of the new method is illustrated by means of applying to truck backer-upper control.