Construction de ressources lexicales sémantiques basée sur le web sémantique : cas de langues Sara-Baguirmiennes du Tchad

Abstract

In the context of digitization and language preservation, constructing semantic lexical networks is a crucial task for natural language processing, research, and the safeguarding of languages. To build these semantic lexical networks, two main approaches are employed: fusion-based and extension-based methods. While effective for well-documented languages, these approaches face challenges when applied to under-resourced languages due to their limited documentation. To address this, we propose a methodology for constructing a multilingual semantic lexical network. This method involves collecting and processing textual corpora, identifying semantic concepts and relations, extracting and representing these concepts and relations, and generating an RDF graph from dictionaries. The expected outcome is a multilingual semantic lexical network for under-resourced languages, facilitating research and preservation efforts while contributing to the appreciation of linguistic and cultural diversity. Additionally, we propose a web platform that allows users to generate RDF files and perform multilingual text searches within these resources using their new representation structure.