Mova French lexicon of the Latin table (forms.tsv)
forms.tsv lists lower-case French word forms with their counts (form<TAB>count; lemmas that never occur as a
form are listed with count 0). It is compiled into langseg (latin::FR_FORMS) and used only to tell French from
other Latin-script languages. It is made by langseg latin-lexicon from these Universal Dependencies 2.18
treebanks:
| treebank | files | licence |
|---|---|---|
| UD French-GSD (Guillaume, de Marneffe and Perrier 2019, from the Google UD treebank of McDonald et al. 2013) | fr_gsd-ud-train.conllu | CC BY-SA 4.0 |
| UD French-ParisStories (Kahane et al. 2021) | fr_parisstories-ud-train.conllu | CC BY-SA 4.0 |
| UD French-Rhapsodie (Lacheret et al. 2014; Bawden et al. 2014) | fr_rhapsodie-ud-train.conllu | CC BY-SA 4.0 |
| UD French-PUD (Zeman et al. 2017) | fr_pud-ud-test.conllu | CC BY-SA 3.0 |
Left out: the GSD dev and test splits (they are the measurement), UD French-ParTUT (CC BY-NC-SA: non-commercial data is for tests only), French-FQB and French-Sequoia (LGPL-LR, a different licence family), French-ALTS (sixteenth-century legal French) and French-PoitevinDIVITAL (Poitevin, a regional language).
The list is a derived work of these treebanks and is distributed under CC BY-SA 4.0 (PUD's 3.0 terms allow it). Treebanks: Universal Dependencies 2.18, Zeman, Nivre et al., LINDAT/CLARIAH-CZ, hdl.handle.net/11234/1-6097 (the release citation as the UD project asks).
This page in Markdown → https://md.mova.elamur.ai/free/en/docs/tools/langseg/data/fr/LICENSE/