LR-POR / PorGram

A Portuguese HPSG Grammar
Other
2 stars 1 forks source link
computational-linguistics grammar-engineering hpsg lingo-grammar-matrix lkb natural-language-processing portuguese portuguese-brazilian

PorGram is an open source computational grammar for Portuguese in the HPSG formalism. It is work in progress. The long term goal is a large coverage grammar for deep syntactic and semantic parsing of unrestricted text.

The grammar is being developed by Leonel Figueiredo de Alencar (@leoalenc), a computational linguist and Professor at the Federal University of Ceará, in collaboration with Alexandre Rademaker (@arademaker), a computer scientist with IBM Research Lab in Rio de Janeiro and Professor at the School of Applied Mathematics of the Getúlio Vargas Foundation (EMAp/FGV).

This effort is part of the research activities of the former as a Visiting Professor at EMAp/FGV. In this first stage of its development, PorGram is being implememnted with the LinGO Grammar Matrix (Bender, Flickinger and Oepen 2002; Bender et al. 2010):

https://matrix.ling.washington.edu/index.html

This system generates an HPSG grammar from specifications provided by the user in a HTML questionnaire. This questionnaire covers a wide range of grammatical phenomena in a large variety of languages. It does not cover, however, the full range of syntactic and morphological properties that need to be encoded in a large coverage grammar. Therefore, additional or alternative specifications will be manually included in the TDL source code generated by the LinGO Grammar Matrix. As an open source project, we invite the community to contribute with the grammar development.