Abstract
Despite being one of the most widely-spoken languages in the world, Portuguese remains a relatively resource-poor language, for which only in recently years NLP tools such as parsers, taggers and (fairly) large corpora have become available. In this work we describe the task of pronominal co-reference annotation and resolution in Portuguese texts, in which we take advantage of information provided by a tagged corpus and a simple annotation tool that has been developed for this purpose. Besides developing some of these basic resources from scratch, our ultimate goal is to investigate the multilingual resolution of Portuguese personal pronouns to improve the accuracy of their translations to both Spanish and English in an underlying MT project. © 2008 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
Cuevas, R. R. M., Honda, W. Y., De Lucena, D. J., Paraboni, I., & Oliveira, P. R. (2008). Portuguese pronoun resolution: Resources and evaluation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4919 LNCS, pp. 344–350). https://doi.org/10.1007/978-3-540-78135-6_29
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
