10.29.2013

Agnes Scaramuzzi Rodrigues

Doutoranda em Linguística Aplicada e Estudos da Linguagem (PUC-SP / LAEL); (2007) Mestra em Linguística Aplicada e Estudos da Linguagem (PUC-SP/LAEL). As duas pesquisas foram inseridas na Linha de Pesquisa Linguagem e Trabalho e contaram com apoio CAPES.

Áreas de interesse: Linguística de Corpus, Linguística Forense, Metáforas Conceptuais, Educação Formal e Corporativa e Ensino de Língua Portuguesa para Estrangeiros e Nativos.

Atuação Profissional: Docente na Uni Sant’ Anna e Faculdade Sumaré.

Membership: Student, PhD.

Corpora: (CELF) Corpora de Estudo para Linguística Forense.

SIGs:

Language Teaching

Metaphor

Forensic Linguistics

Contact: agnes@corpuslg.org

 

09.12.2013

CG3-IDE: A tool for counting linguistic features in texts

The CG3-IDE tool enables users to detect features in texts based on rules and search algorithms defined by the users themselves. It was developed as part of the project 'Dimensions of Register Variation in Brazilian Portuguese' (Fapesp, LAEL, Cepril, PUCSP; investigators: Berber Sardinha, T., Kauffmann, C., Mayer Acunzo, C., from 2010 through 2012).

The tool was developed by Tino Didriksen (GrammarSoft ApS).

It can be downloaded for free here.

 

06.12.2013

Cluster Analysis no SPSS / Cluster Analysis on SPSS

 

Neste vídeo (6.7 MB, MP4), Carol Zuppardo mostra como fazer Cluster Analysis no SPSS, que pode ser usada como parte da Análise Multidimensional.

On this vídeo (6.7 MB, MP4), Carol Zuppardo shows you how to carry out Cluster Analysis on SPSS, which may be useful in a Multidimensional Analysis.

Em português. In Portuguese.

 

05.25.2013

Dimensions of Variation in Brazilian Portuguese (CNPq and Fapesp)

This is a project funded by both the Brazilian Research Council (CNPq) and the São Paulo State Research Foundation (Fapesp). The project took two years to conclude, from November 2010 through October 2012 (CNPq) or February 2011 through January 2013 (Fapesp). The goal was to map the parameters of variation in Brazilian Portuguese through a Multi-Feature, Multi-Dimensional Analysis as proposed by Biber (1988 et seq.). Members of the team:

  • Tony Berber Sardinha
  • Carlos Kauffmann
  • Cristina Mayer Acunzo

Download the Tag Count program for texts tagged with the PALAVRAS parser.

Download the CG3-IDE tool for identifying and counting features in texts.

The corpus was The Brazilian Register Variation Corpus, with 5,644,006 tokens, in 48 different registers.

The dimensions identified are the following:

  1. Oral vs literate discourse/Oralidade versus letramento
  2. Argumentation/Argumentação
  3. Involved versus informational production/Produção com envolvimento versus produção com foco informacional
  4. Procedural discourse/Discurso procedural
  5. Future versus past time orientatioin/Orientação temporal para o futuro versus orientacao temporal para o passado
  6. Reported discourse/Discurso relatado

References:

Berber Sardinha, Tony, Kauffmann, Carlos, & Acunzo, Cristina Mayer. (2011). Dimensions of Variation in Brazilian Portuguese. Paper presented at the Corpus Linguistics Conference  2011, Birmingham, UK.

Berber Sardinha, Tony, Kauffmann, Carlos, & Acunzo, Cristina Mayer. (2011). Dimensions of Variation in Brazilian Portuguese. Paper presented at the XI Encontro de Linguística de Corpus, UFMG, Belo Horizonte, MG.

Berber Sardinha, Tony, Kauffmann, Carlos, & Acunzo, Cristina Mayer. (2012). Dimensions of Variation in Brazilian Portuguese. Talk presented at the English Department, Northern Arizona University, Flagstaff, AZ, USA.

Berber Sardinha, Tony, Kauffmann, Carlos, & Acunzo, Cristina Mayer. (forthcoming). Dimensions of Register Variation in Brazilian Portuguese. In T. Berber Sardinha & T. São Bento Ferreira (Eds.), Working with Portuguese Corpora. London: Bloomsbury/Continuum.

 

 

 

05.25.2013

Tag Count application for corpora tagged with the PALAVRAS parser

This application was developed by Andre Kovacs for the Sao Paulo Catholic University (PUCSP) project 'Dimensions of Variation in Brazilian Portuguese', with funding from Fapesp (Sao Paulo State Research Council). Download the application for Mac (4MB) or in Java (all platforms, 2.1 MB).It will count tags in texts that have been tagged with the PALAVRAS parser and display the counts. Users can customize how counts are run.

 

<< Anterior :: Próximo >>