Communication Dans Un Congrès Année : 2016

Automatically identifying transitions between locutions in dialogue

Résumé

The contribution of this paper is theoretical foundations for dialogical argument mining, as well as initial implementation in software for dialogue processing. Automatically identifying the structure of reasoning from natural language is extremely demanding. Our hypothesis is that the structure of dialogue can yield additional clues as to argument structures that are created and cocreated. Our work has been performed using the MM2012 corpus in OVA+.

Fichier principal
Vignette du fichier
budzynska_15426.pdf (307.37 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
Loading...

Dates et versions

hal-01327096 , version 1 (06-06-2016)

Licence

Identifiants

  • HAL Id : hal-01327096 , version 1

Citer

Katarzyna Budzynska, Mathilde Janier, Juyeon Kang, Barbara Konat, Chris Reed, et al.. Automatically identifying transitions between locutions in dialogue. 1st European Conference on Argumentation (ECA 2015), Jun 2015, Lisbon, Portugal. pp. 1-18. ⟨hal-01327096⟩
260 Consultations
298 Téléchargements

Partager

  • More