Interfacing Sentential and Discourse TAG-based Grammars

Laurence Danlos 1, * Aleksandre Maskharashvili 2 Sylvain Pogodalla 2
* Corresponding author
1 ALPAGE - Analyse Linguistique Profonde à Grande Echelle ; Large-scale deep linguistic processing
Inria Paris-Rocquencourt, UPD7 - Université Paris Diderot - Paris 7
2 SEMAGRAMME - Semantic Analysis of Natural Language
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Tree-Adjoining Grammars (TAG) have been used both for syntactic parsing, with sentential grammars, and for discourse parsing, with discourse grammars. But the modeling of discourse connectives (coordinate conjunctions, subordinate conjunctions , adverbs.. .) in TAG-based formalisms for discourse differ from their modeling in sentential grammars. Because of this mismatch, an intermediate, not TAG-related, processing step is required between the sentential and the discourse processes, both in parsing and in generation. We present a method to smoothly interface sentential and discourse TAG grammars, without using such an intermediate processing step. This method is based on Abstract Categorial Grammars (ACG) and relies on the modularity of the latter. It also provides the possibility, as in D-STAG, to build discourse structures that are direct acyclic graphs (DAG) and not only trees. All the examples may be run and tested with the appropriate software.
Document type :
Conference papers
Complete list of metadatas

Cited literature [37 references]  Display  Hide  Download

https://hal.inria.fr/hal-01328697
Contributor : Sylvain Pogodalla <>
Submitted on : Friday, June 10, 2016 - 9:24:14 AM
Last modification on : Friday, January 4, 2019 - 5:33:24 PM

Identifiers

  • HAL Id : hal-01328697, version 3

Citation

Laurence Danlos, Aleksandre Maskharashvili, Sylvain Pogodalla. Interfacing Sentential and Discourse TAG-based Grammars. Proceedings of the 12th International Workshop on Tree Adjoining Grammars and Related Formalisms (TAG+12), Jun 2016, Düsseldorf, Germany. ⟨hal-01328697v3⟩

Share

Metrics

Record views

351

Files downloads

205