"YKY (Yan King Yin)" <[EMAIL PROTECTED]> wrote:
On 2/28/08, Mark Waser <[EMAIL PROTECTED]> wrote:
>
> >> I think Ben's text mining approach has one big flaw: it can only reason
> >> about existing knowledge, but cannot generate new ideas using words /
> >> concepts
>
> There is a substantial amount of literature that claims that *humans* can't
> generate new ideas de novo either -- and that they can only build up "new"
> ideas from existing pieces.
That's fine, but the way our language builds up new ideas seems to be very
complex, and it makes natural language a bad knowledge representation for AGI.
...
So the meaning of AB depends on the *interactions* of A and B, and it
violates the principle of compositionality -- where the meaning of AB would be
somehow combined from A and B in a *fixed* way.
...
So the simple concept "do X with a knife" can be interpreted in myriad ways
-- it generates new ideas in complex ways.
YKY
agi | Archives | Modify Your Subscription
Although natural language is very difficult to model using computational
knowledge representation systems of some kind, that does not mean it is
impossible, and so far no other method has been significantly successful at it.
All the rational methods use categorization of some kind. It is pretty safe
to say that strictly partitioned categorization is not adequate, and your point
means simplistic compositional methods are not adequate either.
I think that seeing the problem as a compositional problem is important.
I feel that meaning comes from complex relational operations which may
include general meaning vs special meaning, multiple levels of abstraction (for
one example a sentence may include multiple nouns which may refer to objects of
different levels of abstraction or instantiation,) varieties of meaning such as
metaphors and figures of speech, and so on. I also feel that there have to be
complexes of relevant ideas that can act as judgment relations on the subject
being considered.
So the knowledge representation might still be used as long as it was seen as
a part of much more complicated system. In fact, I do not think that
probabilities are even necessary in a sophisticated enough system although they
would be implicit in the filaments of knowledge that the system would produce.
Jim Bromer
---------------------------------
Looking for last minute shopping deals? Find them fast with Yahoo! Search.
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=95818715-a78a9b
Powered by Listbox: http://www.listbox.com
