A model for knowledge representation and manipulation (inference), in knowledge base systems

In artificial intelligence field, choosing the right knowledge representation and manipulation methodologies are considered the most crucial keys of developing a successful knowledge base system. In fact, logic in general and resolution method specifically have been the dominant tools for representing and manipulating knowledge. This led for forming a gap between the knowledge area and the information area, which depends structurally and operationally on set theory in general and on relational algebra in particular, despite the isomorphism exists between the various logics and their set theories counterparts. In this research, we introduced an alternative methodology that has the potential to cover the gap caused by using different mathematical stands in designing knowledge and information systems. This was done, first by conducting a new knowledge representation model that depends structurally on fuzzy and crisp set theories. Then, this model has been used as the base for conducting an inference model that attempts, using a set of algebraic operations and by going through a series of stages, to reach a solution of the problem under study. This reasoning model operates in a manner very close to how, we believe, human experts usually use their knowledge, taking into consideration the speed and accuracy as much as the problem allows. Furthermore, this unified knowledge and inference model was verified on an expert system for medical diagnosis, and its success was proved through experiments on selected patient samples that were taken under the supervision of the domain expert, whom approved the system findings.