[seek-kr-sms] OWL Inference APIs (was Re: SMS stuff)

[Serguei Krivov]

>I generally agree here: it would be really useful to know what exact 
>expressibility is needed in a DL for SEEK (or for ecological
ontologies) 
>... but I think that at this point it isn't clear.  It might be nice to

 
Shawn,
It seems we are all agree that we need to explore the expressive power
of existing DL constructs and see what we need and what we may omit. To
contribute to this end I am attaching the summary of all useful DL
constructs I came across ever! (BTW: do you know any tool that convert
math expression from pdf to any editable format? )

>Actually, here is an interesting paper (that I haven't read yet) that 
>introduces S-Box reasoning for spatial relationships:

>http://kogs-www.informatik.uni-hamburg.de/~haarslev/publications/qr->97
web.pdf

My understanding is that this paper presents earlier efforts of Haarslev
and Mo:ller which they abandoned now. In the page 5 they talk about role
forming predicate operator as alternative to what they call S-Boxes. It
delivers about the same expressive power. In a later paper :
http://kogs-www.informatik.uni-hamburg.de/~haarslev/publications/jlc99.p
df

they presented decidable DL ALCRP(D) that uses this construct to do
spatial reasoning.

BTW- these are just the same guys who develop RACER. Haarslev is running
project on DL and spatial reasoning:
http://kogs-www.informatik.uni-hamburg.de/%7Edls/

 May be we should just ask them what they are planning to do with RACER
in this regard???? Though it is unfortunate that RACER is not free
source.


Best Regards,
serguei
-------------- next part --------------
A non-text attachment was scrubbed...
Name: On Expressive power.doc
Type: application/msword
Size: 467456 bytes
Desc: not available
Url : http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/seek-kr-sms/attachments/20040401/add25c94/OnExpressivepower.doc