Re: Humanscale cognitive Natural Language entities by 2005...

From: CYMM (cymm@trinidad.net)
Date: Sun Jul 23 2000 - 04:17:03 MDT


Hi Damien,

Unfortunately, the idea of deep structure as an underlying object based
semantic modelling of the universe is used in AI.

It is derived from the Chomskian term, I guess; in that it is independent of
the syntax of the language.

You can see such ideas in Brachman ; and Schank & Reiger. It, unfortunately
perhaps, has some value in computational linguistics... especially the
semantic modelling.

Still I'm not so sure that the Chomskian idea of deep structure is
"obsolete" or far from what I'm talking about.. most of the linguists &
psycholinguistics people I've met use the term... or at least understand
it's significance...

The multifarious equivalent representations of a particular meaning were in
the Chomskian scheme of things gottemn bytransforming to the surface
structure.

The concept is a very powerful one in building good smart machines.

alain

...see Brachman in "On the epistemological status of semantic networks" in
Readings in Knowledge Representation...I don't think he uses the term. And
Roger Schank's conceptual dependency theory in "inference & the computer
understanding of NL"



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:30:07 MST