Acquiring semantics from structured corpora to enrich an existing lexicon

Núria Gala (1) , Véronique Rey

1 LIF - Laboratoire d'informatique Fondamentale de Marseille - UMR 6166

Lexical and semantic information is capital in linguistic resources, be it for language learning purposes or for NLP applications. However, this kind of information is very difficult to collect either manually or automatically. The difficulties come from the nature of the information (what do we mean by 'semantics'? how is the 'meaning' put into words?) and from the resources themselves (how are the 'semantics' formalised and displayed?). In this paper we present a methodology to automatically acquire semantics from structured corpora (machine-readable dictionaries and free encyclopaedias on the web). This information is used to enrich an existing lexical database for constructional families of words in modern French.

Mots clés: automatic acquisition; semantic information; morpho-phonology; constructional families; lexical database

Domaines: Linguistique

Núria Gala, Véronique Rey. Acquiring semantics from structured corpora to enrich an existing lexicon. Electronic lexicography in the 21st century: new applications for new users (eLEX-2009), Oct 2009, Louvain-la-Neuve, Belgium. ⟨hal-03198427⟩