Mainly as a primer ontology / knowledge representation data set for an AGI to 
work with.
  Having a number of facts known without having to be typed in about many 
frames and connections between frames gives an AGI a good booster to start with.

  Taken a simple set of common words in a house chair, table, sock, closet etc, 
a house agi bot could get a feel for objects it would expect to find in a 
house, and what locations to look for say a sock, and properties of a sock, 
without having to have that information typed in from a human user.
  Then that information would be updated thru experience, and with a human 
trainer working with an embodied (probably virtual) agi.

The novels gave a really good data set that reinforced the factoids extracted 
and was a bit more world-knowledge common sense than other extraction projects 
using Wall Street Journal, or a subset of the web as hole, and removed much of 
the junk data.

James


Vladimir Nesov <[EMAIL PROTECTED]> wrote: On Dec 11, 2007 7:26 PM, James 
Ratcliff  wrote:
> Here's a basic abstract I did last year I think:
>
> http://www.falazar.com/AI/AAAI05_Student_Abtract_James_Ratcliff.pdf
>
> Would like to work with others on a full fledged Reprensentation system that
> could use these kind of techniques....
> I hacked this together by myself, so I know a real team could put this kind
> of stuff to much better use.
>
>
>
> James

Do you have any particular path in mind to put this kind of thing to
work? Finding patterns is fine, and somewhat inevitable, but what are
those ontologies good for, and why?

-- 
Vladimir Nesov                            mailto:[EMAIL PROTECTED]

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?&;



_______________________________________
James Ratcliff - http://falazar.com
Looking for something...
       
---------------------------------
Looking for last minute shopping deals?  Find them fast with Yahoo! Search.

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=75372394-eb4d01

Reply via email to