UOS-conveners: (01)
I think that the case presented by Adam Pease and the response of
Nicola Guarino are a very good example of the kind of examination of
the relations among the upper ontologies that could, when resolved,
provide either translation axioms or a common subset ontology. This
case is a good example in part because it treats an important and
common relation, which should be in any upper ontology, and it also
elucidates some of the nuances that should be made clear in any upper
ontology that we would recommend to the public for general use. (02)
If it were possible to get funding for a project to create a common
subset ontology, this kind of analysis will be needed, and will, I
think lead at the very minimum to a clear and comprehensive
documentation describing the meanings of the ontology elements (Types,
Relations, Axioms) in sufficient detail to eliminate all significant
ambiguity and permit those elements to be used in the same sense by
everyone. I think that it also shows that, a useful common subset may
be practical to build, provided that making some minor additions or
changes to the existing upper ontologies were acceptable to the
custodians. (03)
In the specific case of SUMO "subProcess" and DOLCE "Temporal Part", I
think Adam is correct that considering only the single axiom he
presented for SUMO subProcess with the DOLCE axiom that axiomatized
Temporal Part does not provide enough of a view of the relations
between these two semantic relations, **but** other axioms in SUMO do
indicate a very close relation between the two semantic relations in
the two ontologies. It appears to me that SUMO "subProcess" and DOLCE
"temporal part" are closely related (the latter may be a subrelation of
the former), and that clarifying the relation between them could show
where useful additions or additional constraining axioms could be
helpful, in the common subset and possibly also in the parent upper
ontologies. I think that Adam is also correct in believing that
finding the precise relations will be a non-trivial exercise for many
such cases. But I think that the exercise will be worthwhile. (04)
I will present some analysis on this specific issue in an attached file
to illustrate the point, and will also refer to the related OpenCyc
relation in that analysis. (05)
I would conclude from examining this case that the SUMO "subProcess"
and DOLCE "Temporal Part" (PT or P.T) are different mostly in that a
"subProcess" could include a spatial part, which is a separate relation
in DOLCE. So the DOLCE "Temporal Part" appears to be a specialization
of the "subProcess" relation. These relations differ in the way the
additional implications they entail are expressed, but are closely
enough related that they both could be represented by relations in a
common subset ontology that could be translated into relations in each
of the upper ontologies, provided that (1) a more general
"subPerdurant" relation were added to DOLCE which subsumed both
"Temporal Part" and "Spatial Part" relations on perdurants, and which
would be more closely equivalent to "subProcess"; and (2) every
Perdurant in DOLCE were equivalent to a Process in SUMO. Likewise, the
OpenCyc "subSituations" appears to be identical to "subProcess", but
that is not specified in axioms, rather it is implied in the OpenCyc
documentation. Again, to be identical, a Situation in OpenCyc would
have to be equivalent to a Process in SUMO and a Perdurant in DOLCE. (06)
For that reason, I believe that "subProcess", "subSituation" and
"Temporal Part" could all be accommodated in a common subset ontology,
depending on how closely identified SUMO "Process", OpenCyc "Situation"
and DOLCE "Perdurant" are. (07)
It is unclear whether DOLCE accommodates a specialized relation
specifying spatiotemporal parts of Perdurants. If not, this should be
added to make a closer match with SUMO, OpenCyc, BFO, and ISO 15926. (08)
I do hope it is possible to get funding for this kind of study. (09)
Pat (010)
Patrick Cassidy
MITRE Corporation
260 Industrial Way
Eatontown, NJ 07724
Mail Stop: MNJE
Phone: 732-578-6340
Cell: 908-565-4053
Fax: 732-578-6012
Email: pcassidy@xxxxxxxxx (011)
-----Original Message-----
From: uos-convene-bounces@xxxxxxxxxxxxxxxx
[mailto:uos-convene-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Adam Pease
Sent: Wednesday, March 08, 2006 6:00 PM
To: Upper Ontology Summit convention
Subject: Re: [uos-convene]: Relating ontologies (012)
Nicola,
I'm not sure how the DOLCE axiom could be inaccurate. I just
copied it from a DOLCE paper. Maybe it's just from an older
version. But of course, that's not really the point. The point is
that it's hard to merge formal ontologies. That you can't understand
one SUMO relation from one example axiom (among many) that uses it is
part of the point. You would have to look at all the axioms that
involve the term to appreciate its meaning, and then try to align
that with all of DOLCE's relevant axioms. That's hard, to say the
least.
Whether we call them axioms or definitions doesn't matter. If
there are rules or other statements that use the terms, and they
aren't shared by both models, they would need to be reconciled.
SUMO's subProcess is not a primitive. It is defined as is every
other term in SUMO. See
<http://sigma.ontologyportal.org:4010/sigma/Browse.jsp?kb=SUMO&lang=en&;
term=subProcess> (013)
Adam (014)
At 02:49 PM 3/8/2006, Nicola Guarino wrote:
>Dear Adam,
>
> I am very confused by your example.
>
>1. First of all, your transcription of the DOLCE axiom is inaccurate,
>since the "subset" relation you mention in the DOLCE example is
>indeed the temporal inclusion relation, defined in terms of temporal
>locations of perdurants. We have no subset relation in the DOLCE
>vocabulary.
>
>2. You are comparing a SUMO *axiom* concerning processes with a DOLCE
>*definition* concerning the notion of temporal part. I guess that a
>more interesting comparison could be between the SUMO relation
>"subProcess" (which I understand is taken as primitive, i.e. not
>defined) and the DOLCE relation "TemporalPart" (defined in terms of
>perdurant, part, and the above mentioned temporal inclusion).
>
>3. I don't know whether "SubProcess" is suitably constrained in SUMO.
>Intuitively, from the name, I understand a subprocess should be some
>how a "part" of a process. Now two questions arise: a) is there a
>formal relationship between SubProcess and Part in SUMO? b) if yes,
>does any part of a process count as a subprocess? The DOLCE
>definition clarifies these two questions, saying that a temporal part
>X of a process Y is a part which is "temporally maximal", in the
>sense that all parts of Y which are temporally included in X are also
>parts of X. So a non-temporally maximal part of a process is not a
>temporal part.
>
>4. We may conclude that, *IF* SUMO has equivalents of the DOLCE
>notions of parthood and temporal inclusion, then the DOLCE definition
>of temporal part could be used to better clarify the SUMO notion of
>subprocess. In practice, limiting ourselves to this very simple
>example, a suitable alignment with DOLCE may result in a more precise
>ontology, in the sense that some non-intended SUMO models may be
>excluded thanks to the DOLCE axiomatization. [NOTE: I am using the
>term "precise" in a very technical sense - see my work on precision,
>coverage and accuracy as dimensions for comparing and evaluating
>ontologies: "Toward a Formal Evaluation of Ontology Quality." IEEE
>Intelligent Systems 19, no. 4 (2004): 78-79.]
>
>5. However, I only focused on the SUMO subProcess relation in this
>discussion. Considering the full axiom you reported results in more
>puzzlement, since I cannot grasp its meaning: apparently, it just
>says that every subprocess has a time. Not very informative...
>
>My conclusion is that a careful comparison between SUMO and DOLCE
>concerning the relationship between the mereological structure of
>processes and their temporal location could actually result in a
>better understanding of these notions. I am sure that such improved
>understanding could be of benefit for SUMO users, as well as for
>DOLCE users willing to to comunicate with SUMO users.
>
>Best,
>
>Nicola
>
>
>On Mar 8, 2006, at 10:10 PM, Adam Pease wrote:
>
>>Hi John,
>> The example I used was of SUMO's Process vs. DOLCE's Perdurant.
>>They cover a similar semantic need, but the details of the formal
>>definitions, and then all the connections to other definitions are
>>so complex and intertwined it seems clear to me that the return on
>>investment for merging isn't there. It's much easier to pick one.
>>Trying to merge formal ontologies seems to me to be harder even
>>than creating a new ontology from scratch.
>>
>>------------------------------------
>>
>> Mapping (hard)
>>
>>- SUMO:Process
>>
>>(=>
>> (and
>> (instance ?PROC Process)
>> (subProcess ?SUBPROC ?PROC))
>> (exists (?TIME)
>> (time ?SUBPROC ?TIME)))
>>
>>- DOLCE:Perdurant
>>
>>TemporalPart(x, y) =df perdurant(x) ^ Part(x, y) ^ forall z((Part
>>(z, y) ^ z subset x) -> Part(z, x)
>>
>>-These are just some of many axioms in each ontology
>>
>>--------------------------------------
>>
>>
>>Adam
>>
>>At 10:01 AM 3/8/2006, bateman@xxxxxxxxxxxxx wrote:
>>
>>>Adam,
>>>
>>>you mentioned just now in the telecon some overheads that
>>>you used yesterday (?) to illustrate irreparable
>>>incompatibility between SUMO and DOLCE. Since you were
>>>last in Bremen 2-3 years ago we have been working more or less
>>>continuously on issues of relating deeply axiomatised
>>>ontologies. I would be interested in seeing the
>>>examples that you used so that we could consider
>>>how we would be going about relating these incompatibilities
>>>with our kinds of tools. This might make some of the
>>>discussion more concrete when it comes to what may or
>>>may not come out of the exercise of relating ontologies
>>>and also help relate to the other initiatives and
>>>actions in this direction.
>>>
>>>Could you send a pointer to the overheads?
>>>Best,
>>>John B.
>>>
>>> _________________________________________________________________
>>>Message Archives: http://ontolog.cim3.net/forum/uos-convene/
>>>To Post: mailto:uos-convene@xxxxxxxxxxxxxxxx
>>>Community Portal: http://ontolog.cim3.net/
>>>Shared Files: http://ontolog.cim3.net/file/work/
>>>UpperOntologySummit/uos-convene/
>>>Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?
UpperOntologySummit
>>
>>----------------------------
>>Adam Pease
>>http://www.ontologyportal.org - Free ontologies and tools
>>
>>
>>_________________________________________________________________
>>Message Archives: http://ontolog.cim3.net/forum/uos-convene/
>>To Post: mailto:uos-convene@xxxxxxxxxxxxxxxx
>>Community Portal: http://ontolog.cim3.net/
>>Shared Files:
>>http://ontolog.cim3.net/file/work/UpperOntologySummit/ uos-convene/
>>Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?
UpperOntologySummit
>
>
>
>
>----------------------------------------------------------------------
-- -----
>Nicola Guarino
>Co-Editor in Chief, Applied Ontology (IOS Press)
>Head, Laboratory for Applied Ontology (LOA), ISTC-CNR
>Institute for Cognitive Sciences and Technologies
>National Research Council
>Via Solteri, 38
>I-38100 Trento
>
>phone: +39 0461 828486
>secretary: +39 0461 436641
>fax: +39 0461 435344
>email: guarino@xxxxxxxxxx
>web site: http://www.loa-cnr.it
>
>
>_________________________________________________________________
>Message Archives: http://ontolog.cim3.net/forum/uos-convene/
>To Post: mailto:uos-convene@xxxxxxxxxxxxxxxx
>Community Portal: http://ontolog.cim3.net/
>Shared Files:
>http://ontolog.cim3.net/file/work/UpperOntologySummit/uos-convene/
>Community Wiki:
http://ontolog.cim3.net/cgi-bin/wiki.pl?UpperOntologySummit (015)
----------------------------
Adam Pease
http://www.ontologyportal.org - Free ontologies and tools (016)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/uos-convene/
To Post: mailto:uos-convene@xxxxxxxxxxxxxxxx
Community Portal: http://ontolog.cim3.net/
Shared Files:
http://ontolog.cim3.net/file/work/UpperOntologySummit/uos-convene/
Community Wiki:
http://ontolog.cim3.net/cgi-bin/wiki.pl?UpperOntologySummit (017)
subProcess.doc
Description: subProcess.doc
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/uos-convene/
To Post: mailto:uos-convene@xxxxxxxxxxxxxxxx
Community Portal: http://ontolog.cim3.net/
Shared Files: http://ontolog.cim3.net/file/work/UpperOntologySummit/uos-convene/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?UpperOntologySummit (01)
|