Yes, but that is all old stuff, which hasn't
worked much better than the simpler Elizae of that
day. I think we should consider a very limited
syntactic approach and a much deeper semantic one. (01)
The point of my post was that using semantics to
drive parsing, with a much more limited role for
parsing, might yield better results than we have
gotten so far, including ATNs and the various
parser approaches. (02)
Of course Cyc and its offspring have been
considered "semantic" but that hasn't been all
that much better, and it certainly hasn't led to
widespread use of Q&A bots. (03)
Perhaps it is time to completely rethink the
approaches of the last half century. (04)
-Rich (05)
Sincerely,
Rich Cooper
EnglishLogicKernel.com
Rich AT EnglishLogicKernel DOT com
9 4 9 \ 5 2 5 - 5 7 1 2 (06)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On
Behalf Of John F Sowa
Sent: Friday, March 01, 2013 7:40 AM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum] Improved Elizae being
used in Call Centers (07)
On 3/1/2013 2:51 AM, Rich Cooper wrote:
> From the bio, [Chetan Dube] was a Math Prof at
NYU with a specialty
> in finite state automata before going
commercial... (08)
> So I started thinking about FSA parsers, perhaps
using a Turing-like stack (09)
If you allow finite-state machines to call one
other, you get much
more expressive power. If they can make direct or
indirect recursive
calls, you get the ability to express context-free
grammars. Such
things are sometimes called recursive transition
nets. (010)
If you add the option of storing and accessing
data from global
storage (the storage is sometimes called
registers, and the data
is often called features), you get some
context-sensitive capability.
If your features are sufficiently complex, you get
the full power
of a Turing machine. (011)
A recursive transition net with features is
sometimes called
an augmented transition net. (012)
In any case, finite state machines + feature
structures with support
for recursion are widely used in computational
linguistics. Ron Kaplan
and his group at Xerox PARC developed them in
great detail as a formal
system, but they can also be used very informally
(i.e., typical hacks). (013)
And Kaplan's group left Xerox to form PowerSet,
which was absorbed
into Microsoft's Bing. (014)
Summary: There is a continuum from Eliza-like
hacks to highly
formal syntax and semantics. Variations of the
same kinds of
technology can support any point along the
continuum. (015)
John (016)
__________________________________________________
_______________
Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr:
http://ontolog.cim3.net/mailman/listinfo/ontolog-f
orum/
Unsubscribe:
mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePa
ge#nid1J (017)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (018)
|