How Could Veins Speed Up The Process Of Discourse Parsing

Elena Mitocariu, Daniel Anechitei, Dan Cristea

Abstract
In this paper we propose a method of reducing the search space of a discourse parsing process, while keeping unaffected its capacity to generate cohesive and coherent tree structures. The parsing method uses Veins Theory (VT), by developing incrementally a forest of parallel discourse trees, evaluating them on cohesion and coherence criteria and keeping only the most promising structures to go on with at each step. The incremental development is constrained by two general principles, well known in discourse parsing: sequentiality of the terminal nodes and attachment restricted to the right frontier. A set of formulas rooted on VT helps to guess the most promising nodes of the right frontier where an attachment can be made, thus avoiding an exhaustive generation of the whole search space and in the same time maximizing the coherence of the discourse structures. We report good results of applying this approach, representing a significant improvement in discourse parsing process.
Anthology ID:
L14-1571
Volume:
Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)
Month:
May
Year:
2014
Address:
Reykjavik, Iceland
Editors:
Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, Stelios Piperidis
Venue:
LREC
SIG:
Publisher:
European Language Resources Association (ELRA)
Note:
Pages:
2871–2878
Language:
URL:
http://www.lrec-conf.org/proceedings/lrec2014/pdf/73_Paper.pdf
DOI:
Bibkey:
Cite (ACL):
Elena Mitocariu, Daniel Anechitei, and Dan Cristea. 2014. How Could Veins Speed Up The Process Of Discourse Parsing. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14), pages 2871–2878, Reykjavik, Iceland. European Language Resources Association (ELRA).
Cite (Informal):
How Could Veins Speed Up The Process Of Discourse Parsing (Mitocariu et al., LREC 2014)
Copy Citation:
PDF:
http://www.lrec-conf.org/proceedings/lrec2014/pdf/73_Paper.pdf
PDF Search