Rob (01)
glad you like the ideas, (02)
About dynamic hierarchy, yes too, in the same way. We need to be able
to insert new 'classters' as the knowledge boundaries may need to be
shifted around (03)
I think the degree to which a hierarchy should be 'fixed' should be
user defined, and by that I don't mean that individual users within a
team adjust their own hierarchy to their personal taste (that would
result in an incoherent model) I mean that depending on the task at
hand, a hierarchy should be more or less fixable or a purpose for a
given time. (04)
Was there a specific problem being addressed? (05)
pdm (06)
On Feb 7, 2008 12:32 PM, Rob Freeman <lists@xxxxxxxxxxxxxxxxxxx> wrote:
> Hi Paola,
>
> Dynamic aggregation sounds like a good term. Aggregations speak of
> "collections", which I think are a good way to handle meaning. And I
> believe the collections must be able to change (with context) so I
> like dynamic too.
>
> Semantic clusters also strikes me as the right way to look at the
> problem. Define your concepts using of collections of terms (so they
> can change dynamically.)
>
> Where I have reservations is with the idea of organizing these
> clusters according to some fixed hierarchy. I think the hierarchy
> needs to be able to change too, not just your path within it.
>
> A dynamic hierarchy needn't be too difficult. I haven't done it for
> semantic classes, but I looked at finding hierarchies for "dynamically
> changing" grammatical classes with some success. Perhaps you could
> approach dynamic hierarchies for semantic classes in the same way.
>
> -Rob
>
> On Feb 6, 2008 8:09 PM, <paola.dimaio@xxxxxxxxx> wrote:
> > no, it means I havent worked it out yet :-)
> >
> > but yes, thats definitely a possibility
> >
> > let's think in terms of 'semantic clusters' (terms/keywords/concepts)
> > that can be organized
> > hierarchically following a classic class structure, however, can be
> > viewed/accessed following different paths, this is the idea of a
> > 'tagging' and other work being done elsewhere
> >
> > 'dynamic aggregation' I think its the term
> >
> > what do you think?
> >
> > On Feb 6, 2008 6:58 PM, Rob Freeman <lists@xxxxxxxxxxxxxxxxxxx> wrote:
> > > Hi Paola,
> > >
> > > Do you mean you assume the classes you find will not be linear/will
> > > vary with context?
> > >
> > > -Rob
>
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
> (07)
--
Paola Di Maio
School of IT
www.mfu.ac.th
********************************************* (08)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|