ontolog-forum
[Top] [All Lists]

[ontolog-forum] Fwd: owl and scalability

To: Ian Horrocks <ian.horrocks@xxxxxxxxxxxxxxx>, "Zhuk, Yefim (Jeff)" <Yefim.Zhuk@xxxxxxxxxxxxx>, "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Peter Yim <peter.yim@xxxxxxxx>
Date: Fri, 23 Jul 2010 06:16:14 -0700
Message-id: <AANLkTin0_B009sRtp68FNniKVem9GbT-_M6kZkxmDLxp@xxxxxxxxxxxxxx>
Thank you, Ian. ... Let's put this conversation onto the discussion
list to share it with the community.    (01)

Jeff, please check out our membership details (see:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J ) in case
you aren't already an Ontolog member (and is therefore not subscribed
to the [ontolog-forum] list yet.)    (02)

Regards.  =ppy
--    (03)


---------- Forwarded message ----------
From: Ian Horrocks <ian.horrocks@xxxxxxxxxxxxxxx>
Date: Fri, Jul 23, 2010 at 2:56 AM
Subject: Re: owl and scalability
To: "Zhuk, Yefim (Jeff)" <Yefim.Zhuk@xxxxxxxxxxxxx>
Cc: Peter Yim <peter.yim@xxxxxxxx>    (04)


Quick answer -- if you need more please come along next week.    (05)

OWL ontologies can be split into multiple files. From a
formal/semantic POV this makes absolutely no difference. Effects on
performance will be very much tool dependent. If you can really
identify a subset of the ontology that is (for example) sufficient to
answer queries at run time, then this may well improve performance
with some tools. Identifying such subsets is highly non-trivial in
general (naive syntactic techniques don't work in general), but there
are now well understood techniques that can achieve semantic
modularisation, and even some tool support -- see, e.g., [1].    (06)

Regards,
Ian    (07)

[1] 
http://www.comlab.ox.ac.uk/people/ian.horrocks/Publications/download/2008/CHKS08a.pdf
---    (08)


On 23 Jul 2010, at 05:01, Peter Yim wrote:
> Hello Jeff,
>
>
> Thank you for the message.
>
> You are probably aware that Professor Horrocks (who gave today's talk
> on "Scalable Ontology-Based Information Systems") will be championing
> two more sessions (on Thu 2010.07.29 & 2010.08.05, i.e. on the next
> two coming Thursdays) on "OWL 2." The Jul-29 session will cover the
> OWL 2 standard and specifications, and the Aug-5 sessions will be on
> Tools and Applications. He has invited a panel of experts to join him
> at the Aug-5 session too. See:
> http://ontolog.cim3.net/forum/ontolog-forum/2010-07/msg00028.html
>
> I would, therefore, strongly suggest that you come join us at these
> two sessions. There will be plenty of time for discussion, and a lot
> of expertise around. You might bring up your questions and suggested
> approach for discussion then ... your question would be especially
> appropriate for the Aug-5 session.
>
>
> Regards. =ppy
>
> Peter P. Yim
> --    (09)


> On Thu, Jul 22, 2010 at 12:11 PM, Zhuk, Yefim <Yefim.Zhuk@xxxxxxxxxxxxx> 
>wrote:
>> Hi Peter!
>>
>>
>>
>> Thank you so much for facilitating this event (7/22, Scalable Ontology-Based
>> Information Systems).
>>
>>
>>
>> A question: OWL can be split into multiple files. What are the best
>> practices to use this feature for better performance.
>>
>> For example, Id think of taking into a separate file a specific sub-domain
>> that is a focus of run-time query, while keeping the rest in another file
>> and have some pre-compiled (design-time created) things that would help at
>> run-time processing. This balance might vary from task to task.
>>
>>
>>
>> Any practices like this or similar and better?
>>
>>
>>
>> Thank you,
>>
>>
>>
>> Jeff    (010)

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: 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 join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (011)

<Prev in Thread] Current Thread [Next in Thread>