Ivandré Paraboni

University of São Paulo - USP

School of Arts, Sciences and Humanities - EACH

me2.jpg                                                        
Artificial Intelligence Research Group - GrIA         

Av. Arlindo Bettio, 1000 - Ermelino Matarazzo                        

03828-000 São Paulo – Brazil

    

Background

 

Research

 

Affiliations

 

Publications

Background

I obtained a PhD degree in Computer Science from the Information Technology Research Institute (ITRI), Brighton, UK, in July 2003. My work concerns Natural Language Processing (NLP) research, and since August 2005 I am a lecturer at the University of São Paulo (first at ICMC and now at EACH). I was also a visiting post-doctoral fellow at the University of Aberdeen in Scotland in 2011-2012.

Research Interests

I am interested in Human Language Technology in a broad sense, including both theoretical issues such as reference phenomena (particularly, the semantic content of referring expressions) and their role in NLP applications. My work has focused on the combination of rule-based and statistical methods for Natural Language Generation, addressing the issues of referring expressions generation, content planning and surface realisation.  I also share a number of interests with my colleagues at NILC regarding Machine Translation systems, statistical approaches to NLP, document classification and information retrieval for web-based applications.

A list of publications can be found here.

Funded Research Projects

·         (Aug 2011 - Feb 2012) Referring Expressions Generation in Interactive 3D Domains (CAPES grant 6415/10-5).
This project concerns the extension of our previous work on the use of logically redundant information in referring expressions generation as a means to facilitate search in large and/or complex domain structures such as those addressed in the GIVE Project.

 

·        (Sept 2009 – Aug 2011) Hybrid Models of Surface Realisation. (FAPESP grant 2009/08499-9).
This project applies the traditional 2-stage generation approach introduced by I. Langkilde and others to the problem of text generation for morphologically-rich languages such as Portuguese, combining standard statistical language models with corpus-based knowledge. (overview)

 

·         (Nov 2007 – Oct 2009) Resolution and Interpretation of Referring Expressions in Text Generation.
(CNPq grant 484015/2007 9).

 

·         (Nov 2006 – Oct 2008) Building a Parallel Corpus with Annotated Co-reference Chains for Statistical Machine Translation. (FAPESP grant 2006/03941-7)

 

·         (Sept 1998 -  July 2003) Generating references in hierarchical domains: the case of document deixis. PhD project at the University of Brighton (CNPq grant 200123/98-0) (abs) (thesis contents)

 

(top)

 

Affiliations

ACL – The Association for Computational Linguistics

SIGGEN - ACL Special Interest Group on Natural Language Generation

Registry of Latin American Researchers in NLP and Computational Linguistics

 

(top)

 

Locations of visitors to this page 

Barely maintained by Ivandré Paraboni.  Last updated 13/02/2012.