Papers by Laurent Kevers
Towards a Corsican Basic Language Resource Kit (2020.lrec-1)
Copied to clipboard
| Challenge: | a roadmap has been set out for the development of a basic language resource kit for the Corsican language . the goal is to improve the availability of resources and tools for the language based on the Banque de Données Langue Corse project . |
| Approach: | a team of researchers from univ-corse is developing a basic language resource kit for the corsican language . they aim to collect corpora, set up a concordancer, set-up language detection tool, build an electronic dictionary and add a part-of-speech tagger . |
| Outcome: | the goal is to improve the availability of resources and tools for the Corsican language . the roadmap sets out the actions to be undertaken: collection of corpora, setting up of a concordancer, language detection tool, electronic dictionary and part-of-speech tagger. |
The ParCoLab Parallel Corpus and Its Extension to Four Regional Languages of France (2024.lrec-main)
Copied to clipboard
Dejan Stosic, Saša Marjanović, Delphine Bernhard, Myriam Bras, Laurent Kevers, Stella Retali-Medori, Marianne Vergez-Couret, Carole Werner
| Challenge: | Parallel corpora are scarce for most of the world's language pairs. |
| Approach: | They propose to extend ParCoLab with a parallel corpus for Alsatian, Corsican, Occitan and Poitevin-Saintongeais. |
| Outcome: | The proposed corpus contains more than 20k tokens per regional language. |
Agettivu, Aggitivu o Aghjettivu? POS Tagging Corsican Dialects (2024.lrec-main)
Copied to clipboard
| Challenge: | a series of experiments towards POS tagging Corsican are presented . POS tags are used to tag a less-resourced language spoken in corsica . |
| Approach: | They present a series of experiments towards POS tagging Corsican . they first contribute to the first gold standard POS-tagged corpus for Corsica . |
| Outcome: | The proposed model is the first POS tagger for Corsican, reaching an accuracy of 93.38%. |