Section: New Results

Towards a French FrameNet

Participants : Marie-Hélène Candito, Marianne Djemaa, Benoît Sagot, Éric Villemonte de La Clergerie, Laurence Danlos.

The ASFALDA project (https://sites.google.com/site/anrasfalda/ ) is a three-year project which started in October 2012, with the objective of building semantic resources (generalizations over predicates and over the semantic arguments of predicates) and a corresponding semantic analyzer for French. We chose to build on the work resulting from the FrameNet project [47] , (https://framenet.icsi.berkeley.edu/ ) which provides a structured set of prototypical situations, called frames, along with a semantic characterization of the participants of these situations (called frame elements, FEs). The resulting resources will consist of :

1. a French lexicon in which lexical units are associated to FrameNet frames,

2. a semantic annotation layer added on top of existing syntactic French treebanks

3. and a frame-based semantic analyzer, focused on joint models for syntactic and semantic analysis.

In the first year of the project, we focused on the first of these objectives. A team of 10 active members, from Alpage, the Laboratoire de Linguistique Formelle (LLF), the MELODI team (IRIT - Toulouse) and the CEA-List partners achieved :

• the delimitation and adaptation to French of a set of FrameNet frames, in order to cover a set of specific notional domains (commercial transaction, communication, cognitive positions, judgment/evaluation, temporal relations, spatial position, causality).

• and the semi-automatic construction of a French lexicon in which French lexical units are associated with frames

The current resource contains 110 frames, and roughly 2500 lexical units / frame pairs. The next phase consists in automatic pre-annotation of semantic annotations, that will serve as basis for the manual validation phase.

Note that a publication describing the project and these first achievements shall be presented at the LREC 2014 conference.