Abstract—In this paper, we propose an ontology enhancement framework to accommodate imprecise concepts and their inter-relations mined from text documents. The proposed framework is modeled as a fuzzy ontology structure to represent concept descriptor as a fuzzy relation which encodes the degree of a property value using a fuzzy membership function. In our application, the fuzzy membership function is determined through text mining. Other than concept descriptors, the inter-concept relations in the ontology are also associated a fuzzy strength. The strength of association between two concepts determines the degree of association between the concepts. The fuzzy ontology with fuzzy concepts and fuzzy relations is an extension of the domain ontology with crisp concepts and relations which is more suitable to describe the domain knowledge for solving uncertainty reasoning problems. The applicability of the fuzzy ontology structure in mining and managing imprecise knowledge from biomedical text documents has been thoroughly experimented. The fuzzy ontology can later be used for information curation to answer imprecise queries posed at multiple levels of specificities along the underlying ontology.

Index Terms—ontology engineering, ontology enhancement, fuzzy ontology, imprecise concept, text mining, knowledge management  

Cite: Muhammad Abulaish, "An Ontology Enhancement Framework to Accommodate Imprecise Concepts and Relations," Journal of Emerging Technologies in Web Intelligence, Vol. 1, No. 1, pp. 22-36, August 2009.