OntoLex validation

Dear all,

in todays morphology telco, we once again stumbled upon the question how  
to validate OntoLex data (also cf. https://book.validatingrdf.com).

There are basically four possibilities, but some won't work with certain  
vocabulary elements in OntoLex

- RDFS validation: Possible only if we do not use disjunctive domains such  
as in in http://www.w3.org/ns/lemon/decomp#constituent
- OWL validation: Should work with rdf:Containers (rdf:_1, as mentioned as  
a possibility for decomp), but not with rdf:Collections (rdf:List,  
()-notation in Turtle). What bothers me a bit is that we model closed  
(non-extensible) collections in decomp, but use a vocabulary that is  
designed to model open (extensible) collections.
- SHACL: Not tested yet. Does anyone have experiences?
- special-purpose validator: I remember something like this used to exist  
for Monnet, but this is ancient history ...

Basically, there are three questions here:
1) Did anyone look into OWL validation for OntoLex-lemon so far?
2) Would anyone be interested in looking into SHACL or alternative  
validation strategies and/or provide an OntoLex validator?
3) Shall we return to RDFS validation and eliminate disjunctions (by  
introducing abstract classes)?

Option 1) is probably the most elementary choice, but because of the open  
world assumption, this is a weak validation only, and a complex one.
Option 2) is a lot of work.
Option 3) is a backward-compatible extension, but requires the  
introduction of novel vocabulary into OntoLex core.

Maybe something to be discussed in Leipzig.

Best,
Christian
-- 
Prof. Dr. Christian Chiarcos
Applied Computational Linguistics
Johann Wolfgang Goethe Universität Frankfurt a. M.
60054 Frankfurt am Main, Germany

office: Robert-Mayer-Str. 11-15, #107
mail: chiarcos@informatik.uni-frankfurt.de
web: http://acoli.cs.uni-frankfurt.de
tel: +49-(0)69-798-22463
fax: +49-(0)69-798-28334

Received on Tuesday, 9 April 2019 12:24:21 UTC