Recognition of alternation paraphrases: a robust and exhaustive symbolic approach

Marilisa Amoia 1 Claire Gardent 1
1 LANGUE ET DIALOGUE - Human-machine dialogue with a significant language component
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : In this paper we show how by incorporating linguistic knowledge in a shallow parser like XIP, it is possible to build a robust semantic parser which can cope with paraphrastic constructions involving alternations and/or lexical synonymy. The robustness of the parser is dependent on the amount of linguistic knowledge at disposal. For the moment, we have incorporated the VerbNet verb lexicon in our parser, so that it can recognise verb synonymy and meaning preserving alternations of declarative sentences. In the future, we want to extend the parser taking into consideration grammatical variations of declarative sentences and more complex cases of paraphrastic constructions, such as nominalisations and noun phrase-adjective substitution.
Document type :
Conference papers
Complete list of metadatas

https://hal.inria.fr/inria-00000249
Contributor : Claire Gardent <>
Submitted on : Thursday, October 6, 2005 - 4:17:42 PM
Last modification on : Thursday, January 11, 2018 - 6:19:48 AM

Identifiers

  • HAL Id : inria-00000249, version 1

Collections

Citation

Marilisa Amoia, Claire Gardent. Recognition of alternation paraphrases: a robust and exhaustive symbolic approach. Knowledge and Reasoning for Answering Questions - KRAQ'05, 2005, Edinburgh/Scotland. ⟨inria-00000249⟩

Share

Metrics

Record views

90