PARNT : A statistic based approach to extract non-taxonomic relationships of ontologies from text

摘要

Learning Non-Taxonomic Relationships is a subfieldof Ontology learning that aims at automating theextraction of these relationships from text. This articleproposes PARNT, a novel approach that supports ontologyengineers in extracting these elements from corpora of plainEnglish. PARNT is parametrized, extensible and uses originalsolutions that help to achieve better results when compared toother techniques for extracting non-taxonomic relationshipsfrom ontology concepts and English text. To evaluate thePARNT effectiveness, a comparative experiment with anotherstate of the art technique was conducted.