This repository contains AMRScien-Br, a corpus of scientific news texts in Brazilian Portuguese annotated with Abstract Meaning Representation - AMR. The corpus is composed by two hundred sentences which come from the FAPESP corpus (Aziz and Specia, 2011), a Portuguese-English paralell corpus (pt-en) of scientific news texts belonging to the Brazilian magazine Pesquisa FAPESP Online (https://revistapesquisa.fapesp.br).
For more information about what is AMR and its specific notations, we indicate the AMR guidelines (https://github.com/amrisi/amr-guidelines).
This corpus is distributed under the CC-BY-NC-SA license.
AMRScien-Br follows a standard notation to ease the reading of files. The corpus file contains multiple sentences, each with some metainformation, which starts with a hashtag followed by double colons (# ::) and a keyword (id, snt). Then, the AMR graph representation in the PENMAN notation is written. An example is shown below:
# ::id 16
# ::snt Os otimistas entendem que a crise é passageira.
(u1 / entender-01
:ARG0 (o0 / otimista)
:ARG1 (t3 / passageiro
:domain (c2 / crise)))
A blank line separates each sentence.
For more details about the annotation guidelines, please see here.