GRAMPAL

GRAMPAL was originally developed as a morphological processor of Spanish for written texts (Moreno 1991, Moreno y Goñi 1995). In order to annotate C-ORAL-ROM, new modules were specifically developed for spoken Spanish: a tokenizer, desambiguation modules and a recognizer of unknown words (Moreno y Guirao 2003). These modules were developed for Spanish with the aim of adapting the morphological processor to the spoken language features. GRAMPAL lexicons comprises around 50.000 entries of stems, endings and multiwords. All the dictionary entries, except dependent morphemes and multiwords, have morphological and Part-of-Speech information, including lemmas. Also, GRAMPAL contains 239 prefixes which form new words without a change of category.

Publications

• MORENO SANDOVAL, A. & GOÑI, J. M. "A morphological model and processor for Spanish implemented in Prolog". En: ALPUENTE, M. & SESSA, M.I. (ed.). Proceedings of Joint Conference on Declarative Programming. Salerno: Palladio Editrice, 1995. p. 321-331.
• MORENO, A. & GUIRAO, J. M. "Tagging a spontaneous speech corpus of Spanish". En: Proceedings of the International Conference on Recent Advances in Natural Language Processing.). Borovets, Bulgaria, 2003. p. 292-296.
• MORENO, A. y GUIRAO, J.M." Morpho-syntactic Tagging of the Spanish C-ORAL-ROM Corpus: Methodology, Tools and Evaluation.", in Spoken Language Corpus and Linguistic Informatics, John Benjamins, 2006.

Consult online