Approche évolutive des notions de base pour une représentation thématique des connaissances générales
Abstract
In the field of Natural Language Processing, in order to work out a thematic representation system of general knowledge, methods leaning on thesaurus are used since about fifteen years. A thesaurus consists of a set of concepts which define a generating system of a vector space modelling general knowledge. These concepts, often organized in an arborescent hierarchy, constitute a fundamental, but completely fixed tool. Even if the concepts evolve (we think for example of the technical fields), a thesaurus can evolve as for him only at the time of a particularly heavy process, because requiring the collaboration of human experts. After we detailed the characteristics which a generating system of the vector space of knowledge modelling must have, we define the "basic notions". Those, whose construction is based initially on the concepts of a thesaurus, constitute another generating system of this vector space. We approach the determination of the acceptions expressing the basic notions, which naturally leads us to ask the question of their number. Lastly, we clarify how, being freed from the concepts of the thesaurus, the basic notions evolve by an iterative process progressivel with the analysis of new texts.
Domains
Other
Fichier principal
alain.joubert_document_Poster_TALN_2006.pdf (184.46 Ko)
Télécharger le fichier
alain.joubert_resume_ResumeAbstract.pdf (27.68 Ko)
Télécharger le fichier
Loading...