DiSeg is the first discourse segmenter for Spanish using the framework
of the Rhetorical Structure Theory (Mann and Thompson, 1988)
based on lexical and syntactic rules.
If you want to test it, you can use this demo
(enter your text in Spanish with utf8 enconding):
Credits
You can also try DiSeg on these corpus.
The source code is availabble under GPL here as a shell/perl script
running on linux and relying on FreeLing.
An online API is also available, contact eric.sanjuan@univ-avignon.fr for details.
If you like DiSeg, please cite us as follows:
- da Cunha Iria, SanJuan Eric, Torres Moreno Juan-Manuel, Lloberes Marina, Castelló Irenen (2012): DiSeg 1.0: The first system for Spanish discourse segmentation. Expert Syst. Appl. 39(2): 1671-1678
- da Cunha, Iria; SanJuan, Eric; Torres-Moreno, Juan-Manuel; Lloberes, Marina; Castellón, Irene (2010). Discourse Segmentation for Spanish based on Shallow Parsin . Lecture Notes in Computer Science 6437. 13-23. Berlín: Springer. ISSN 0302-9743
- da Cunha, Iria; SanJuan, Eric; Torres-Moreno, Juan-Manuel; Lloberes, Marina; Castellón, Irene (2010). DiSeg: Un segmentador discursivo automatico para el españo. Procesamiento del Lenguaje Natural 45. ISSN 1135-5948
- da Cunha, Iria; Torres-Moreno, Juan-Manuel (2010). Automatic Discourse Segmentation: Review and Perspectives. En Proceedings of the International Workshop on African Human LanguagesTechnologies. Djibouti, África.
©2010 IULA / LIA / UB