Self-Tuning Fuzzy Rule Bases with Belief Structure

Abstract

A fuzzy rule-based evidential reasoning (FURBER) approach has been proposed recently, where a fuzzy rule-base designed on the basis of a belief structure (called a belief rule base) forms a basis in the inference mechanism of FURBER. This kind of rule-base with both subjective and analytical elements may be difficult to build in particular as the system increases in complexity. In this paper, a learning method for optimally training the elements of the belief rule base and other knowledge representation parameters in FURBER is proposed. This process is formulated as a nonlinear multi-objective function to minimize the differences between the output of a belief rule base and given data. The optimization problem is solved using the optimization tool provided in MATLAB. A numerical example is provided to demonstrate how the method can be implemented.