[plt-scheme] inference.plt
I'm certainly not opposed to taking it in that direction. The current
implementation is based on a simple rules syntax that is very Lisp/Scheme
oriented. I have been mainly looking at interaction with standard PLT
Scheme representations: lists, associations lists, vectors, structures, and
(at some point) objects. I have thought about extended that to XML data and
string processing (using regular expressions) as well. I haven't given much
though toward OWL and/or RDF support. I would probably look at how to add
that kind of functionality on top of the inference engine, rather than as an
integral part of it. But, that is just a gut feel that might not be the
right approach.
I've been working on getting the underlying rule network functionality down.
For me, the important aspects of this functionality are support for both
data-oriented and goal-oriented inferencing (in place), assumption
processing (probably out this weekend), and truth maintenance (future work).
I'm not sure if there are other GOFAI (Good Old Fashioned AI) people on the
list, but inputs on what is/would be desirable in an inference engine for
PLT Scheme are certainly welcome and encouraged. It would be helpful to get
other inputs.
The current implementation reflect my own needs more than anything else.
Those lie in being able to interact with the simuation collection or other
similar tools / applications. I tend to use rule-based systems as just a
part of an overall application - for example, in modeling human
decision-making processes in larger simulaion contexts.
Doug
-----Original Message-----
From: plt-scheme-bounces at list.cs.brown.edu on behalf of Hans
Oesterholt-Dijkema
Sent: Thu 7/20/2006 5:04 PM
To: mzscheme Mailing List
Subject: [plt-scheme] inference.plt
I saw a new PLaneT package 'inference.plt'. Will it be able
to handle OWL/RDF in the future?
--Hans
_________________________________________________
For list-related administrative tasks:
http://list.cs.brown.edu/mailman/listinfo/plt-scheme