The OpenWordNet-PT and its connection to SUMO Ontology
|Alexandre Rademaker||Getulio Vargas Foundation, Brazil||[Home Page]|
Notice: Hosted by Rodrigo de Salvo Braz
Date: Tuesday, August 7th 2012 at 4:00pm
Location: EJ228 (SRI E building) (Directions)
We wish to create a foundation for computer understanding of Portuguese text. Two components of such a foundation are a lexicon of words and a formal ontology to support reasoning. This is the approach of the Global WordNet Grid, which relates lexicons from different languages to a single language independent ontology, the Suggested Upper Merged Ontology (SUMO).
We aim to make this objective concrete, and show incremental progress, by addressing an important archive of documents about history and politics in Brazil, from The Center of Brazilian Contemporary History (CPDOC) at Fundacao Getulio Vargas (FGV). We began in 2010 to process and create formal ontological representations of the contents of these documents, to help make the archives more useful through machine understanding and machine reasoning. More generally, this project is about the creation of a foundation for computational linguistics work in Portuguese similar to that which exists in English. The absence of such resources has been an obstacle to much potential computational linguistics research in Brazil.
Our team has begun work on a Brazilian Wordnet, the OpenWordNet-PT, freely available at https://github.com/arademaker/wordnet-br.
This OpenWordNet-PT would be very valuable to linguists and computer scientist interested in representing knowledge obtained from texts. We will discuss briefly why we want a Brazilian Portuguese WordNet and how we are going about getting one. We aim to improve the OpenWordNet-PT, linked to the logical representations of the SUMO formal ontology and to expand the ontology to handle cultural and historical concepts particular to Portuguese and the CPDOC corpus.
In April 2010 I obtained my PhD at PUC-Rio. My advisor was Edward Hermann Haeusler. My main research areas are Description Logics and Proof Theory. Currently, I am a professor at Applied Math School at Getulio Vargas Foundation.
During my PhD I had two outstanding experiences as International fellow. The first one at MS Research, between April and July of 2008 where I worked with Leonardo de Moura and Nikolaj Bjørner. The second one was at SRI International, between February and July of 2009, where I worked with Natarajan Shankar.
Please arrive at least 10 minutes early as you will need to sign in by following instructions by the lobby phone at Building E. (or call Wilma Lenz at 650 859 4904, or Vicenta at Lopez at 650 859 5750). SRI is located at 333 Ravenswood Avenue in Menlo Park. Visitors may park in the parking lots off Fourth Street. Detailed directions to SRI, as well as maps, are available from the Visiting AIC web page. There are two entrances to SRI International located on Ravenswood Ave. Please check the Builing E entrance signage.