http://ontologydesignpatterns.org/index.php?action=history&feed=atom&title=Reviews%3AAldoGangemi_about_ConceptGroup 2026-06-01T00:55:59Z Revision history for this page on the wiki MediaWiki 1.25.6 http://ontologydesignpatterns.org/index.php?title=Reviews:AldoGangemi_about_ConceptGroup&diff=5815&oldid=prev 2009-09-12T11:01:55Z

<p>New page: {{Content OP Proposal Review Template |CreationDate=2009/9/12 |SubmittedBy=AldoGangemi |ContentOPUnderReview=ConceptGroup |RevisionID=5807 |Score=0 - needs major revision |ReviewSummary=Th...</p> <p><b>New page</b></p><div>{{Content OP Proposal Review Template<br /> |CreationDate=2009/9/12<br /> |SubmittedBy=AldoGangemi<br /> |ContentOPUnderReview=ConceptGroup<br /> |RevisionID=5807<br /> |Score=0 - needs major revision<br /> |ReviewSummary=The pattern tries to provide an OWL representation of a construct used to talk about groups of concepts or fragments of thesauri.<br /> I think the pattern is a valid contribution, but needs major revisions in documentation, axiomatization, and relation to other patterns.<br /> |ReviewConfidence=High<br /> |ReviewProblems=The lack of a concrete example makes the pattern difficult to understand. From the technical viewpoint, there are some problems:<br /> <br /> 1) the object properties lack inverses: this is a bad practice; inGroup lacks an inverse at all, while BT is not made inverse of NT, and subGroup of superGroup<br /> 2) the reification of BT-NT, and subGroup-superGroup mught be due to the need of creating inverse object properties: in this case, this is a bad practice. If reifications are needed for other reasons, this is not clear<br /> 3) the structuringAssociationType object property seems used to link concept groups to the reification of BT-NT. Why is is needed? The use case behind this object property is unclear<br /> |ReviewRelevance=Fair<br /> |ReviewBestPractice=The pattern allows to represent groups of concepts, in that being a special case of the CollectionEntity pattern (Collection &lt;- Group, Thing &lt;- Concept, memberOf &lt;- inGroup).<br /> Moreover, it seems to encode also:<br /> a) a subgroup relation (could be kind of partof for collections)<br /> b) the BT relation (already existing in SKOS core, but could also be kind of specialization for concepts, see also Specialization and Classification patterns)<br /> c) a &quot;membership restriction&quot; class that is basically synonym to Group<br /> I think this pattern can be represented in a simplier way (see Problems above)<br /> |ReviewReusability=Good, if the problems are solved<br /> |ReviewRelations=CollectionEntity can be specialized for the core part of this pattern.<br /> Classification and Specialization are also related.<br /> SKOS (potentially a content pattern has also overlaps).<br /> |ReviewUnderstandability=Low<br /> |ReviewClearProblem=Must be explained better<br /> |ReviewClearRelevance=Must be explained better<br /> |ReviewFigures=Ok<br /> |ReviewMissing=Concrete example. I also suggest to load the annotations in the OWL file, not just on the portal.<br /> }}</div>

AldoGangemi