ProbFuzzOnto: A Fuzzy Ontology-Driven Uncertainty Approach Using Fuzzy Bayesian Networks

Abstract The need to deal with uncertain semantics is rising in importance in most of the important technology trends, and consequently, many proposals have emerged as solutions in recent years. Fuzzy ontologies were proposed to remedy the limitations of standard ontologies using fuzzy logic to deal with vague and imprecise knowledge. Nevertheless, fuzzy ontologies cannot deal with probabilistic knowledge which is an important characteristic of most real-world applications. This paper proposes a novel solution that aims at enhancing the knowledge representation and reasoning in fuzzy ontologies. Indeed, the proposed solution is a probabilistic extension of fuzzy ontologies with Fuzzy Bayesian Networks (FBN) that we named Probabilistic Fuzzy Ontologies (ProbFuzzOnto). It takes into account vague, imprecise, and probabilistic knowledge simultaneously. Moreover, this paper proposes a process to guide ontology engineers step by step in building ProbFuzzOnto. Also, it provides reasoning algorithms to drive implicit knowledge by utilizing explicit knowledge stored in a fuzzy ontology based on fuzzy Bayesian inference. To show the usefulness of the proposed solution, a case study in Renal Cancer is presented.