No inverse relation for similar/near_synonym (i.e. near_antonym)

[simongray]

I noticed that similar is basically the GWA name for the relation called near_synonym elsewhere.

However, what about near_antonym? I have multiple near_antonym relations in the old Danish wordnet dataset and I am not sure how to represent them when there's only similar/near_synonym.

There's a selection of antonym relations available in the form of...

• Antonym (antonym)
• Gradable Antonym (anto_gradable)
• Simple Antonym (anto_simple)
• Converse Antonym (anto_converse)

... but neither of them seem to describe near antonyms.

I guess the general antonym relation is most applicable, but I think using it would result in some information loss...? I would be grateful to hear any better suggestions you may have.

[jmccrae]

Any relation can be specified by means of the other property and the type annotation.

This property could be added if other wordnet projects also require it.

[simongray]

So other + type is how you are supposed to model "foreign" relations in WN-LMF, right?

From reading the example snippet it is not entirely clear to me how I would use it. The example has adc:type (does dc: refer to Dublin Core?), but the "emotion" relation type isn't mentioned anywhere else on the page so I'm a bit unsure what the example actually expresses. I guess "dc:type" is actually a relation name, but its namespace is somehow implied?

How would you express a foreign relation in WN-LMF that is explicit? Using a full URI?

Rght now I am modelling our data using RDF rathe than WN-LMF, but I want to match the relations in GWADOC as closely as possible. I guess in this case I should just express it with either our own custom relation or the EuroWordNet relation, rather than use other + type?

[jmccrae]

There isn't really an option to use an external URL in the LMF serialization of the schema. In RDF, you can just define your own relation of course.