Re: SUO: Multi-Source Ontology (MSO) Draft Ballot Question
> We're getting closer to the release of this ballot ..
> Please post any further comments.
I have not yet received an answer from Christiane Fellbaum.
In addition to the browsers for the MSO, a static file (in
the FT notation) for the main taxonomy is at
http://www.webkb.org/theKB_categs.ft and, since it is a very
big file, I have just isolated the top-level (the part above or
unrelated to WordNet) into http://www.webkb.org/top_categs.ft
While I integrate the SUMO, I'll often modify these files.
I am behind schedule but encountered no problem yet except may be
for the fact that in SUMO, SelfConnectedObject (and its subtype
CorpuscularObject) is both subtype of Object (hence something
physical or at least spatial) and has subtypes like Language and
SymbolicString, the equivalents of which are not physical/spatial
in my ontology. From my current understanding, it seems adequate
to have the following (which involves no subtype link between
sumo#self_connected_object and sumo#object):
pm#entity (^something that can be "involved" in a situation^)
> {(pm#spatial_object pm#nonspatial_object)}
{(pm#undivisible_entity pm#divisible_entity)}
sumo#self_connected_object ...
pm#spatial_object = sumo#object;
pm#divisible_entity > {sumo#corpuscular_object sumo#substance};
sumo#self_connected_object > sumo#corpuscular_object;
...
pm#description_medium //an indirect subtype of pm#entity
> sumo#language pm#abstract_data_type wn#structure
wn#communication wn#language_unit wn#symbolic_representation,
= sumo#content_bearing_object ;
If this is seen as inadequate, I'll change it.
> The following page article has some info about
> a CE that is translated to KIF:
> http://www.clp.ox.ac.uk/people/staff/pulman/pdfpapers/cl_kr.pdf
Yet another alternative: FE (Formalized-English), parsed by WebKB-1
but not yet WebKB-2); comparisons to KIF and other notations in
http://www.webkb.org/doc/papers/iccs02/
Philippe