On Sun, 1 Dec 2002, Alan Grimes wrote:

> > My dream is that someday Cyc will be able to reason about its own
> > behavior as UML state machines and then supercompile (drive the state
> > machine) into optimal machine code for the expected 64-bit CPUs.  I
> > have a UML state machine interpreter working now in java and can
> > represent all the UML objects as Cyc assertions.
>
> In principle, that's a great idea.
>
> Unfortunately CyC is an idiot savant at this point. I would give Cyc
> less than 1% chance of ever getting to the stage you describe. =\ The 1%
> chance being that the focus of the project switch to hard research on
> the Translation Problem and totally abandon the data-entry activities.

In a couple of weeks the director of the Darpa Information Processing
Technology Office, Ron Brachman, is visiting Austin to solicit ideas for
new Darpa programs.  We hope to spend all day with him addressing our
solutions to the problem you refer to above.  I have asked for a 45 minute
slot to describe my ideas which have been described in this forum.

Not to dispute the overall point, I would like to clarify Cycorp's current
knowledge entry activity.  Up to now we have not actually performed much
data entry, rather we have encoded a broad range of commonsense classes
(we call Collections) and relatinships (we call predicates) in symbolic
logic (which we call CycL).  The downside is that Cyc is presently passive
and can answer deductive queries about its contents.  The upside is that
the Cyc commonsense ontology (structure of Collections, Predicates, facts
and rules) can be extended into any domain you might want to describe.
Experiments have shown that Cyc is at the point where 75% of the high and
mid level concepts required for a new domain (for example high school
chemistry) already exist.  I was able to add vocabulary to represent UML
state machines rather easily, for example.  We have a team of
computational linguists who have added the vocabulary to make Cyc able to
represent lexical concepts.

In 2003, our data entry activities will be emphasized as a result of our
participation in Darpa's Total Information Awarenesss program for which we
will construct a Terrorism Knowledge Base, containing all the open-source
terrorist individuals, organizations and events that we and our
sub-contractor experts can input.  It is hoped that the TKB will prove
useful, by answering interesting paramertized questions, for US
defense and intelligence communities.

-Steve

-- 
===========================================================
Stephen L. Reed                  phone:  512.342.4036
Cycorp, Suite 100                  fax:  512.342.4040
3721 Executive Center Drive      email:  [EMAIL PROTECTED]
Austin, TX 78731                   web:  http://www.cyc.com
         download OpenCyc at http://www.opencyc.org
===========================================================

-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]

Reply via email to