Section: New Results

Linguistic Knowledge Representation: the Unit Graphs Formalism

Participants : Fabien Gandon, Maxime Lefrançois.

As any community of interest, linguists produce knowledge. Generic needs arise with such produced knowledge: how to represent it, how to manipulate it, how to share it, how to query it, and how to reason with it. To answer these needs is the goal of the knowledge representation (KR) domain. Existing KR formalisms such as the Semantic Web formalisms are standard solutions, and their specialization to the linguistic domain is under active development. Yet, the description logic behind the OWL formalism fails to represent how the meaning of words combine to build up the meaning of sentences. To tackle this specific problem, we introduced the new so-called Unit Graphs KR framework that is portable to existing KR standards but that introduces its own formal logic. UGs are defined over a UG-support that contains: i) a hierarchy of unit types which is strongly driven by the actantial (from action) structure of unit types, ii) a hierarchy of circumstantial symbols, and iii) a set of unit identifiers. On these foundational concepts, we defined UGs, justified the introduction of a deep-semantic representation level for the Meaning–Text Theory, we represented lexicographic definitions of lexical units, and we introduced two formal semantics: one based on UGs closure and homomorphism, and one based on model theoretic semantics. The UGs formalism has been the object of 6 publications in [42] , [43] , [44] , [45] , [46] , [47] .

A prototype of a GUI of an editor of formal dictionary definitions aimed at lexicographers was developed based on the formalism of Units Graphs and on Meaning-Text Theory. The development of the GUI was preceded by the elaboration of scenarios of how users would interact with Units Graphs objects. It was followed by user tests of the GUI with actual lexicographers unfolding the scenarios. This work is reported in [46] .