top of page

Morfetik morphological resource enhancement and recovery

Morfetik is a morphological dictionary of contemporary French, including an XML and MySQL flexion engine, consultation interfaces and NLP exploitation tools. Morfetik allows you to obtain, for any French lexical entry, all its forms (plural of nouns, feminine and plural of adjectives, conjugated forms of verbs), or, conversely, to identify the word (the base form) corresponding to any inflected form. It currently includes almost 800,000 distinct forms. The resource can either be consulted via a search engine, or downloaded in XML format (under LGPL-LR license). Finally, the interface is connected to those of Wikipedia and FranceTerme1.

The lexical census has drawn on numerous lexicographical sources, but the inventory is not yet complete. It is currently being pursued by adding neologisms and integrating new specialties, but also by refining the description, such as: the appropriate treatment of variable-gender nouns(un(e) élève; un instituteur, une institutrice); the development of frequency markers (possibilitý of taking into account statistical indicators of presence in corpus); the clarification of the link between graphic variants; the finer treatment of relations between forms and uses; etc.


Post: Blog2_Post
bottom of page