Toward a concrete framework for intelligent linguistic modeling
Arya Aghili Ashtiani, Mohammad Bagher Menhaj · 2013
In this paper, the modified fuzzy relational modeling structure is configured in a harmonious manner to obtain some useful properties with both theoretical and practical significance. An appropriate derivative-based iterative identification algorithm is presented for the proposed structure. The proposed model configuration along with the proposed identification algorithm constitute a powerful modeling framework which can be used in several applications. In the context of intelligent modeling, the resulting modeling framework, while preserving a high modeling capability, alleviate the lack of analyzability of the model somehow. It also prevents the existence of hard conflicts between the rules in the rule-base. Furthermore, high relative errors in output computations are prevented by using the proposed scheme.