Section: New Results

Common Basic Resources

Participants : Maxime Amblard, Clément Beysson, Philippe de Groote, Bruno Guillaume, Guy Perrier, Sylvain Pogodalla, Karën Fort.

Corpus Annotation

The Universal Dependencies project (UD) aims at building a syntactic dependency scheme which allows for similar analyses for several different languages. Bruno Guillaume and Guy Perrier are active in the UD community, and participate to the development and the improvement of the French data in this international initiative. Bruno Guillaume converted a new French treebank into UD: the French Question Bank (FQB), developped by Djamé Seddah and Marie Candito  [35]. With the conversion system described in [2], the corpus UD_French-FQB was introduced in version 2.4 of UD in May 2019.

Bruno Guillaume, Marie-Catherine de Marneffe (Ohio State University, Columbus, Ohio, USA) and Guy Perrier improved the consistency of two French corpora annotated with the UD scheme [6]. They improved the annotations of the two French corpora to render them closer to the UD scheme, and evaluated the changes done to the corpora in terms of closeness to the UD scheme as well as of internal corpus consistency.

Bruno Guillaume and Guy Perrier developed and popularized the use of the GREW tool for various language applications and more particularly the pattern matching module Grew-match [22], [26], [17].

SUD is an annotation scheme for syntactic dependency treebanks, that is almost isomorphic to UD (Universal Dependencies). Contrary to UD, it is based on syntactic criteria (favoring functional heads) and the relations are defined on distributional and functional bases. In [14], Kim Gerdes (Sorbonne nouvelle, Paris 3), Bruno Guillaume, Sylvain Kahane (Université Paris Nanterre) and Guy Perrier recalled and specified the general principles underlying SUD, presented the updated set of SUD relations, discussed the central question of Multiword Expressions, and introduced an orthogonal layer of deep-syntactic features converted from the deep-syntactic part of the UD scheme.

FR-FraCas

Maxime Amblard, Clement Beysson, Philippe de Groote, Bruno Guillaume, Sylvain Pogodalla and Karën Fort carried on the development of FR-FraCas, a French version of the FraCas test suite [31] which is an inference test suite, in English, for evaluating the inferential competence of different NLP systems and semantic theories. There currently exists a multilingual version of the resource for Farsi, German, Greek, and Mandarin. Sémagramme completed the first translation into French of the test suite. The latter has been publicly released (https://gitlab.inria.fr/semagramme-public-projects/resources/french-fracas). We also ran an experiment in order to test both the translation and the logical semantics underlying the problems of the test suite. The experiment was run with 18 French native speakers. Such an experiment provides a way of checking the hypotheses made by formal semanticists against the actual semantic capacity of speakers (in the present case, French speakers), and allows us to compare the results we obtained with the ones of similar experiments that have been conducted for other languages [30], [29].