home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group99a.txt
/
000094_icon-group-sender _Fri Apr 9 08:09:21 1999.msg
< prev
next >
Wrap
Internet Message Format
|
2000-09-20
|
2KB
Return-Path: <icon-group-sender>
Received: (from root@localhost)
by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA11954
for icon-group-addresses; Fri, 9 Apr 1999 08:07:19 -0700 (MST)
Message-Id: <199904091507.IAA11954@baskerville.CS.Arizona.EDU>
Date: Thu, 8 Apr 1999 19:21:04 -0500 (CDT)
From: "John C. Paolillo" <johnp@ling.uta.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: Assumption Grammars in Icon?
Cc: johnp@ling.uta.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Dear Icon group --
Icon has a number of features that are related to logic
programming languages (e.g. it allows nondeterministic
depth-first search through automatic backtracking), but
these days in logic programming and NLP-type applications
especially, we are starting to see more and more use of
"linear logic" to help solve certain problems. I don't
have a good definition of linear logic handy, but as I
understand it, it allows one to treat certain kinds of
information as being "consumed" when certain kinds of
conditions are met. This is helpful in many kinds of NLP
tasks, e.g. when you need to match a verb with its arguments
in a language where word order is free (arguments can be in
any order with respect to the verb). I was wondering if
anyone has (or knows of) any ideas about implementing
linear logic like constructs in Icon. My sense is that
very probably this can be done without any modifications
to the language, but I'm a bit unsure as to how it would
be done. Can anyone suggest something (a technique, or
a reference to a paper)?
If you need to see something in order to get a sense of the
kind of thing that linear logic allows, Dahl and Tarau
discuss an implementation of what they call "assumption
grammars" in Tarau's BinProlog, in the following paper:
Assumption Grammars: Parsing as Hypothetical Reasoning
http://clement.info.umoncton.ca/html/JAGS/html.html
(For those that read comp.lang.prolog, this was recently
mentioned there).
Thanks,
John C. Paolillo
UTA Linguistics