ONTOPRIMA:APROTOTYPE FOR AUTOMATING ONTOLOGY POPULATION

Ontology Population supports the process of building ontologies in the complex task of instantiating ontology. Performing this process manually is both expensive and time consuming; this logically leads to attempts of fully or partially automating the process of acquisition and absorption of knowledge in general and the process of Ontology Population in particular. This paper presents OntoPRiMa (Ontology Population for PRIMA Project), a prototype based on a semi-automatic approach for ontology population in the domain of risk management. OntoPRiMa demonstrates the feasibility of the proposed approach in populating generic and small domain ontologies from textual corpus with many potentials and possible instances that are unknown and not realized. The prototype isbased on Natural Language Processing (NLP) techniques for language processing, semantic web techniques (RDFS, RDF, Jena APIs) for knowledge modeling and representation and on domain expert’s intervention for validating extracted instances. Furthermore, this paper presents an experiment for populating the ontology of the PRIMA project using a text corpus which is consist of chemical fact sheets issued by the U.S. Environmental Protection Agency. We have achieved high precision (above 80%) of knowledge acquisition in this experiment.