Advanced

Automatic learning of discourse relations in Swedish using cue phrases

Karlsson, Stefan and Nugues, Pierre LU (2010) 7th International Conference on NLP, IceTAL 2010 In Advances in Natural Language Processing / Lecture Notes in Computer Science, 6233. p.179-184
Abstract
This paper describes experiments to extract discourse relations holding between two text spans in Swedish. We considered three relation types: cause-explanation-evidence (CEV), contrast, and elaboration and we extracted word pairs eliciting these relations. We determined a list of Swedish cue phrases marking explicitly the relations and we learned the word pairs automatically from a corpus of 60 million words. We evaluated the method by building two-way classifiers and we obtained the results: Contrast vs. Other 67.9%, CEV vs. Other 57.7%, and Elaboration vs. Other 52.2%.

The conclusion is that this technique, possibly with improvements or modifications, seems usable to capture discourse relations in Swedish.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
Advances in Natural Language Processing / Lecture Notes in Computer Science,
editor
Loftsson, Hrafn; Rögnvaldsson, Eiríkur; Helgadóttir, Sigrún; ; and
volume
6233
pages
179 - 184
publisher
Springer
conference name
7th International Conference on NLP, IceTAL 2010
external identifiers
  • wos:000289187000021
  • scopus:77956603902
ISSN
0302-9743
1611-3349
ISBN
978-3-642-14769-2
DOI
10.1007/978-3-642-14770-8_21
language
English
LU publication?
yes
id
3b49351a-0d0f-4531-89f0-0e8c17682cd3 (old id 1668785)
date added to LUP
2010-12-10 12:17:50
date last changed
2018-05-29 11:34:14
@inproceedings{3b49351a-0d0f-4531-89f0-0e8c17682cd3,
  abstract     = {This paper describes experiments to extract discourse relations holding between two text spans in Swedish. We considered three relation types: cause-explanation-evidence (CEV), contrast, and elaboration and we extracted word pairs eliciting these relations. We determined a list of Swedish cue phrases marking explicitly the relations and we learned the word pairs automatically from a corpus of 60 million words. We evaluated the method by building two-way classifiers and we obtained the results: Contrast vs. Other 67.9%, CEV vs. Other 57.7%, and Elaboration vs. Other 52.2%. <br/><br>
The conclusion is that this technique, possibly with improvements or modifications, seems usable to capture discourse relations in Swedish.},
  author       = {Karlsson, Stefan and Nugues, Pierre},
  booktitle    = {Advances in Natural Language Processing / Lecture Notes in Computer Science,},
  editor       = {Loftsson, Hrafn and Rögnvaldsson, Eiríkur and Helgadóttir, Sigrún},
  isbn         = {978-3-642-14769-2},
  issn         = {0302-9743},
  language     = {eng},
  pages        = {179--184},
  publisher    = {Springer},
  title        = {Automatic learning of discourse relations in Swedish using cue phrases},
  url          = {http://dx.doi.org/10.1007/978-3-642-14770-8_21},
  volume       = {6233},
  year         = {2010},
}