Morfetik : Une ressource lexicale morphologique extensible et modulaire pour le français
Abstract
Morphological lexical resources, describing the internal structure of words and their inflected
forms, are crucial for natural language processing (NLP) and computational linguistics.
We present MORFETIK, a comprehensive open-source lexical resource for French, capable
of automatically generating and identifying all inflected forms of words (nouns, verbs, adjectives,
phrases, etc.). It offers broad coverage of the contemporary and specialised lexicon, an
extensible and modular architecture, and easy integration with external resources.
We also illustrate its use through two case studies and detail its architecture, showing how
its modularity and interoperability facilitate corpus analysis and the development of NLP tools.