[Edu-sig] CP4E: Programming for Fun => Python for Journalists and Couch Poatatoes?

[Jason Cunliffe]

> What is the (noncommercial) state of the art in automated parsing of
> natural language text?  Is a project like this feasible, or does it
> need a Cyc-like knowledge base to disambiguate words based on context?

Hi Fred

Here are some I am aware of. There's LOTS  more including Java toolkits which
can be used with Jython, and all the cool stuff on the edges of sequencing and
visualization work. BioIT is generating all sorts of tools which can be used for
other disciplines.

POVTalk, which is a Natural Language based 3D scene generator
http://homepages.ihug.co.nz/~synopsis/computing/

Proposal: A Natural Language Programming Toolkit for Python
Edward Loper, Steven Bird July 6, 2002
http://nltk.sourceforge.net/tech/proposal.pdf

The NL Toolkit (NLTK) is a Python package intended to simplify the task of
programming natural language processing systems. Its primary audience is
graduate and undergraduate students studying computational linguistics
http://nltk.sourceforge.net/

PyWordNet is a Python interface to the WordNet database of word meanings and
lexical relationships.  (A lexical relationship is a relationship between words,
such as synonym, antonym, hypernym ("poodle" -> "dog"), and hyponym ("poodle" ->
"dog").
http://pywordnet.sourceforge.net/

./Jason