Back to first pageBack to first page Centre for Artificial Intelligence of UNL
Browse our site

Medieval Portuguese Corpora, automatic POS-tagging and partial parsing of this corpora

Project information
Description

Reuse of tools built for Contemporaneous Portuguese in texts from the XIIIth and XIVth centuries.

Started in September 1996 and was concluded in 1999.

Project

Participating entities: FCSH/UNL, Dep. Computer Science FCT/UNL.

Principal researcher: Francisca Xavier (FCSH/UNL).

Funding

Funding entity: PRAXIS.

CENTRIA

Principal researcher: Gabriel Pereira Lopes.

Results

In 1999, a neural network based POS-tagger was trained and used for POStagging new medieval text; using the grammar of Contemporaneous Portuguese and the parsers constructed, those Medieval Portuguese texts were partially parsed; as a consequence the lexicon of Medieval portuguese was enriched. Methods for extracting multi-word lexical units was used in (Crónica Geral de Espanha). 2 publications were produced.


Centre for Artificial Intelligence of UNL
Departamento de Informática, FCT/UNL
Quinta da Torre 2829-516 CAPARICA - Portugal
Tel. (+351) 21 294 8536 FAX (+351) 21 294 8541

Fundacao_FCT