Comment arpenter sans m\`etre : les scores de r\'esolution de cha\^\ines de cor\'ef\'erences sont-ils des m\'etriques ? (Do the standard scores of evaluation of coreference resolution constitute metrics ?)

no code implementations • JEPTALNRECITAL 2020 • Adam Lion-Bouton, Lo{\"\i}c Grobol, Jean-Yves Antoine, Sylvie Billot, Ana{\"\i}s Lefeuvre-Halftermeyer

Cet article pr{\'e}sente un travail qui consiste {\`a} {\'e}tudier si les scores les plus utilis{\'e}s pour l{'}{\'e}valuation de la r{\'e}solution des cor{\'e}f{\'e}rences constituent des m{\'e}triques de similarit{\'e} normalis{\'e}es.

coreference-resolution

Enhancing the PARSEME Turkish Corpus of Verbal Multiword Expressions

no code implementations • LREC (MWE) 2022 • Yagmur Ozturk, Najet Hadj Mohamed, Adam Lion-Bouton, Agata Savary

We provide an overview of the problems observed in the morphosyntactic annotation of the Turkish PARSEME corpus.

Evaluating Diversity of Multiword Expressions in Annotated Text

no code implementations • COLING 2022 • Adam Lion-Bouton, Yagmur Ozturk, Agata Savary, Jean-Yves Antoine

We apply the validated measures to annotations in 14 languages produced by systems during the PARSEME shared task on automatic identification of multiword expressions and on the gold versions of the corpora.

Lemmatization