↓ Aller au contenu

Séminaires d’équipe

Algorithms for Discontinuous Constituency Parsing

Intervenant·e
Maximin Coavoux (LIG, GETALP)
Date

In this seminar, I will present an overview of my work on discontinuous constituency parsing.
Constituency trees represent the syntactic structure of a sentence by a recursive grouping of tokens into labelled phrases such as NP, PP, VP, etc.
However, standard constituency trees are not adequate to model a number of syntactic phenomena related to word order variations, e.g. long distance extractions.
In contrast, discontinuous constituency trees use crossing branches to model these phenomena.
Discontinuous structures correspond to derivations of mildly context-sensitive formal grammars, such as Linear Context-Free Rewriting Systems (LCFRS), and are therefore much more difficult to parse than projective constituency trees (CFG derivations).

In this talk, I will present efficient data-driven parsing algorithms for discontinuous constituency trees, as well as empirical parsing results on English and German discontinuous parsing.

I will also mention the release of wikiparse: wikipedia automatically annotated with constituency trees, dependency trees and morphological analyses (POS tags, morphological features) for 7 languages.

References: