OntoPop or how to annotate documents and populate ontologies from texts

Autor: Amardeilh, Florence
Přispěvatelé: Langues, logiques, informatiques, cognition (LaLIC), Université Paris-Sorbonne (UP4)-Centre National de la Recherche Scientifique (CNRS), Modèles, Dynamiques, Corpus (MoDyCo), Université Paris Nanterre (UPN)-Centre National de la Recherche Scientifique (CNRS), Mondeca Recherche & Développement (MONDECA R&D), Mondeca, Amardeilh, Florence
Jazyk: angličtina
Rok vydání: 2006
Předmět:
Zdroj: ESWC 2006 Workshop on Mastering the Gap: From Information Extraction to Semantic Representation
ESWC 2006 Workshop on Mastering the Gap: From Information Extraction to Semantic Representation, Jun 2006
Popis: http://sunsite.informatik.rwth-aachen.de/Publications/CEUR-WS//Vol-187/11.pdf; This paper presents an innovative solution for annotating documentsand populating domain ontologies in a single process. In fact, this process canbe viewed as a translation of a linguistic representation of a text into a moreformal representation format, being RDF statements or OWL instances. Thus,we provide the OntoPop methodology for mapping linguistic extractions withconcepts of ontology thanks to knowledge acquisition rules. This methodologyhas been implemented in a platform integrating a IE tool and a commercial ontologyrepository. The platform is currently under evaluation on various realworld applications. Nevertheless, OntoPop already raises interesting issues suchas semantic heterogeneity that are discussed at the end of this article.
Databáze: OpenAIRE