Re: [ontolog-forum] API4KB and diverse ontologies

[Benjamin Grosof <benjamin.grosof@xxxxxxxxx>]

Hi Elisa,

I'd like to talk to you about the technical approach in API4KB, to understand what you're doing and see if I might be able to help including at the design level.

We are building some knowledge interchange capabilities at Coherent, including between Rulelog and OWL but also with other standards.

I have a lot of experience with design and implementation of KB/KRR interoperability, which has been a major part of all the systems I've built and released since the mid 90s. 

Could we schedule a time for next week?  When might be good times for you? 

Also, will you be coming to AAAI (or even RuleML, which is co-located)?  Note I'll be giving a tutorial at AAAI. 

Best,
Benjamin

 

On Fri, Jun 28, 2013 at 12:26 PM, Elisa Kendall <ekendall@xxxxxxxxxxxx> wrote:

John,

One of the primary goals of the APIs is to provide standard interfaces
for constructing hybrid reasoning systems, so that
classification-oriented reasoners can be incorporated in a standard way
with enterprise quality rule engines, predictive analytics, data mining
and other such capabilities to solve real world problems.  I'm also
interested in improving the interoperability of ontology development
using RDFS, OWL, and CL with other modeling tools, for example to
facilitate reasoning about systems engineering and business process models.

The parsers/APIs are not up to snuff sufficiently to support doing this
without a great deal of effort, including custom development that may
not work if the APIs themselves change  (I can say this as having been a
victim of the results more than once).  There are a number of APIs out
there, notably the OWL API, that folks use for development, but it isn't
a standard at this point, and could benefit from commercialization.
Consistent integration of the OWL API and even RDF-based APIs such as
Jena can be challenging.

For the kinds of systems I'm concerned about, including those that would
use extensions of the emerging Financial Industry Business Ontology
(FIBO) for use in banking applications, the ontology(ies) themselves
would be reused across systems, with appropriate mappings from
qualitative to quantitative measures/metrics.  So -- we're not talking
about multiple ontologies in various languages with differing semantics
at all in this case, but ways of integrating the same ontology in
different engines.  Early demonstrations of the potential have been very
well received by the institutions and regulators alike.  But -- the
access methods for the reasoners are insufficient for our needs, which
include integration with service busses, use in REST architectures, and
so on.  Matthew Horridge has said that he would be interested in working
with us to use the API4KBs wrappers together with the OWL API to show
how this could be done in practice, which is one of our critical test
cases.  Other test cases include the Reaction RuleML architecture
(Adrian Paschke, Harold Boley, Davide Sottara, and others), which
supports very interesting integration of rule systems and other kinds of
analyses already, but has not been integrated with description logics
reasoners to date.  Yet another group participating in the work includes
Jeff Pan and his team working on OWL reasoners that simulate some of the
aspects of OWL that can raise performance issues through rewriting,
etc.  This would give us at least two participants with varied OWL
reasoning capabilities and interfaces, and in addition to the RuleML
folks, IBM and TIBCO are also on our voting list and participate from
time to time in the meetings/development activities, with possible
implementations in ILOG and TIBCO Business Events, among other
commercial applications.

Best regards,

Elisa

On 6/28/2013 9:41 AM, John F Sowa wrote:
> Ed,
>
>> But you can't blow hot and cold with the same mouth here.
> I'm blowing ambivalence.
>
> The initial impression I got from Sottara's 2012 slides was that he
> was trying to do something very ambitious that was premature for
> standardization.  But you and Elisa convinced me that he is doing
> something much simpler -- and certainly doable.
>
> My only objection is we already have a solution:  Common Logic is
> a superset of all the logics on their list.  Any or all of the three
> dialects in the 2007 ISO standard or the next revision -- CLIF, CGIF,
> or XCL -- could be used for interchange among the systems on their list.
>
> So what's the point?
>
> John
>
>
>
> _________________________________________________________________
> 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
>
>
>


_________________________________________________________________
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

--

Benjamin Grosof, PhD 
LinkedIn:  http://www.linkedin.com/in/benjamingrosof ; Homepage:  http://www.mit.edu/~bgrosof/ ; Phone:  +01 (206) 232-3214 (business/home).  

_________________________________________________________________
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    (01)