2013-12-04T14:02:35Z
2013-12-04T14:02:35Z
2010-09-01
2013-12-04T14:02:35Z
In this paper we present WRPA, a system for Relational Paraphrase Acquisition from Wikipedia. WRPA extracts paraphrasing patterns that hold a particular relation between two entities taking advantage of Wikipedia structure. What is new in this system is that Wikipedia's exploitation goes beyond infoboxes, reaching itemized information embedded in Wikipedia pages. WRPA is language independent, assuming that there exists Wikipedia and shallow linguistic tools for that particular language, and also independent of the relation addressed.
Artículo
Versión publicada
Inglés
Lingüística computacional; Paràfrasi; Corpus (Lingüística); Tractament del llenguatge natural (Informàtica); Computational linguistics; Paraphrase; Corpora (Linguistics); Natural language processing (Computer science)
Sociedad Española para el Procesamiento del Lenguaje Natural (SEPLN)
Reproducció del document publicat a: http://journal.sepln.org/sepln/ojs/ojs/index.php/pln/article/view/786
Procesamiento del lenguaje natural , 2010, num. 45, p. 11-19
(c) Vila Rigat, Marta et al., 2010