Interval Certitude Rule Base Inference Method using the Evidential Reasoning

Abstract

Development of rule-based systems is an important research area for artificialintelligence and decision making, as rule base is one of the most general purposeforms for expressing human knowledge. In this paper, a new rule-based representationand its inference method based on evidential reasoning are presented based on operationalresearch and fuzzy set theory. In this rule base, the uncertainties of humanknowledge and human judgment are designed with interval certitude degrees whichare embedded in the antecedent terms and consequent terms. The knowledge representationand inference framework offer an improvement of the recently developed rulebase inference method, and the evidential reasoning approach is still applied to therule fusion. It is noteworthy that the uncertainties will be defined and modeled usinginterval certitude degrees. In the end, an illustrative example is provided to illustratethe proposed knowledge representation and inference method as well as demonstrateits effectiveness by comparing with some existing approaches.