ANN: Proxem Antelope, a Natural Language Processing framework for .NET

microsoft.public.dotnet.framework

14 Sep 2006 12:29 AM

François-Régis Chaumartin

September, 2006.

We have the pleasure to announce the new version 0.7 of Proxem Antelope, a
full Natural Language Processing framework for Microsoft .NET 2.0. It can be
freely downloaded on www.proxem.com.

It now includes a full tutorial, a syntactic Web search utility, and many
improvements to its modules:

- Access to many part-of-speech taggers, including the SS Tagger,
- A NP-chunker,
- Access to the Link Grammar dependency parser,
- Access to the Stanford Parser (no Java VM needed!),
- A full lexicon, with rich relations, based on WordNet 2.1 data,
- A syntax / semantic layer, based on VerbNet 1.5 (experimental),
- An anaphora resolver (experimental),
- A space and time features detector (experimental),
- A collocation collapser,
- A sentence splitter for plain text or HTML text,
- A PROLOG interpreter for .NET,
- An Information Extraction module.

The middle-term objective of Antelope (version 1.0, expected in mid-2007) is
to offer a true, easy to use, semantic parser. It will be first applied on
an encyclopedia (a large subset of Wikipedia seems to be a nice target) to
make complex queries possible at a semantic level.

More and more people had been downloading Antelope.
We hope you will give us some feedback about the way you use it!

Sincerely yours,

François-Régis Chaumartin
Proxem Webmaster
www.proxem.com