Re: Re: nature -> "human brain" -> "language terms" ==>> knowledge ?
Rich,
you wrote:
>
> Actually, I've started work on a prototype lattice that uses
> WordNet's noun
> classes. So far, I have a first order Q&A function to move from one node
> in the lattice to another. I think the same thing is needed for
> the verbs,
> but
> haven't gotten that far.
>
> What operational requirements do you see for a lattice-based ontology
> using WordNet concepts?
>
Since any categorisation at the level of linguistics (and so WordNet) is already 'biased' I would suggest going a bit 'lower/deeper' to at least (if you wish to stay at the lingustics level of things) set down core gerunds as the 'whole', out which come the A-X vs TO-X POTENTIALS in meaning. Local context will then actualise (and the mechanism of nominalisation/denominalisation allow for dynamics across the A/TO elements in the dichotomy)
In WordNet if you search for "running" you get (a) a list of "running" categorised as a NOUN and (b) a list of "run" categorised as a VERB. The IDM perspective would put the GERUND format as a level of meaning 'above' the noun/verb dynamic where the derivation out of the gerund is on "A RUN" vs "TO RUN". The fact that in LOCAL grammars nouns/verbs become biased often hides the underlying, but vague, root meanings that the LOCAL context may consider being 'wrong'. E.g.:
A Sowa vs TO Sowa. In the current context of this list, the former is a noun and the latter a process but the latter would be considered 'strange' until we recognise the fact that any noun can be translated into a verb and visa versa - thus TO Sowa would reflect the application of properties/methods to something where the original source is in the behaviour of Sowa. Same thing for A Lofting vs To Lofting (the latter reflect the push for abuse to generate anger and so a competitive aspect that can bring out such things as priorotisation preferences for values over facts; preparedness to 'persevere' in 'difficult' conditions, to see 'past' the fluff etc etc)
The core focus in communications is differentiating/integrating and their occasional mediation and so issues of push-away,pull-along/pull-in,push-along. At the level of IDM's gerunds of blending, bonding, bounding, binding we have core qualities sourced at the neurological level that seed all upper levels. That seeding reflects 'inheritence' and as such upper level differentiations use those seeds together with local refinements to become the seed for the next level.
What you identify as a node, is more a PAIR of potentials, one 'noun' oriented, one 'verb' oriented. The node itself is therefore a a gerund format, a superposition where CONTEXT extracts the 'best fit' element out of that superposition. When you get to the level of not being able to differentiate further then you are at the level of a base element of the dichotomy represented at the previous level.
Indicative here is the use of WAVE or PULSE representations that can allow for many to share the one space. Different languages will use this to varying degrees - thus English will 'demand' more universalisations, more labels, whereas Chinese can be more context-sensitive in requirements to interpret something.
All of these languages are ACTUALISATIONS of the ONE set of POTENTIALS - covered in the IDM template of POTENTIALS. Local context will use ad hoc methods in actualisations and so develop biases in perspectives.
As such, the mapping of POTENTIALS gives us a 'pure form' that is then overlayed with local 'anomolies' and so a map of actuals; and so the ideal lattice pattern gets distorted due to environmental 'pressures'. As we differentiate so we cut and make borders and out pops complexity/chaos - the underlying order is in the set of POTENTIALS that is constant but vague - thus what emerges is still 'seeded' with the basic structures from the realm of potentials.
Chris.
This message was sent through MyMail http://www.mymail.com.au