Design of fuzzy knowledge based systems with multi-set confidence models

In many application domains, the use of AI has been restrained due to the lack of viable models and tools for dealing with evaluations of uncertainties of domain knowledge. One of the feasible alternatives is based on fuzzy algebras of strict monotonic operations with multisets. Nevertheless the implementation of such a model in knowledge engineering tools is a challenging issue since many traditional approaches do not work for such model. In the paper, the multiset-based model of uncertainty is proposed and associated specifics of the construction and validation of such knowledge-based systems is discussed. The underlying model and engineering tools have been used in applications for petroleum engineering, which illustrate the proposed approach.