The Proteus Project
The Proteus Project is a project of research and development in
natural language processing, conducted in the Department of Computer
Science, New York University, under the direction of
Prof. Ralph Grishman.
The basic research is focussed on the automated acquisition of
linguistic knowledge from large corpora. We are developing methods for
the acquisition of probabilistic grammars, semantic relations between
words, and selectional patterns.
This research is tied to applications in four areas:
- information retrieval
(the retrieval of documents from large collections)
- information extraction
(extracting information about particular types of events from free text)
- machine translation
- language modeling for speech recognition
In addition, the project is involved in developing infrastructure in three areas:
The project is supported by grants and contracts from the
Advanced Research Projects Agency (ARPA), the National Science
Foundation, and the Linguistic Data Consortium.