Can't explore thesaurus

[Ismael Olea]

Hi:

I'm working with this thesaurus[1], configured skosmos (it works with others, so installation seems fine) and added the thesaurus to config.ttl:

:tpha a skosmos:Vocabulary, void:Dataset ;
    dc:title "Tesauro de Patrimonio Histórico Andaluz"@es ;
    skosmos:shortName "TPHA";
    dc:subject :cat_general ;
    void:uriSpace "https://www.iaph.es/tesauro#";
    skosmos:language "es";
    skosmos:defaultLanguage "es";
    void:sparqlEndpoint <http://fuseki-cache:80/skosmos/sparql> ;
    skosmos:sparqlGraph <https://www.iaph.es/tesauro#>.

[1] https://juntadeandalucia.es/datosabiertos/portal/iaph/dataset/dataset/6c199ca2-8d2e-4c12-833c-f28233a5251d/resource/38731571-42af-4373-b425-6dae73bacf79/download/tesauro-iaph.rdf

I can load it, but can't explore:

I tried also to skosify it, but the result is the same:

skosify --namespace=https://www.iaph.es/tesauro# --label="Tesauro de Patrimonio Histórico Andaluz" --default-language=es --no-infer --set-modified -i tesauro-iaph.rdf -o 

I am completely new to SKOS, etc. I am missing something probably obvious?

--

Ismael.

[Ismael Olea]

On Friday, September 15, 2023 at 2:03:27 PM UTC+2 Ismael Olea wrote:

I tried also to skosify it, but the result is the same:

skosify --namespace=https://www.iaph.es/tesauro# --label="Tesauro de Patrimonio Histórico Andaluz" --default-language=es --no-infer --set-modified -i tesauro-iaph.rdf -o 

ups, to -o really is -o output.rdf.

 

[Osma Suominen]

Hello Ismael,

sorry for the delayed reply. I checked your thesaurus and it seems to be
in SKOS XL format, with concept labels only expressed using SKOS XL.
Skosmos requires that also SKOS Core labels are present. For more
detailed requirements, see here:
https://github.com/NatLibFi/Skosmos/wiki/Data-Model

You can convert SKOS XL to SKOS Core e.g. using a SPARQL query. There is
also a conversion tool among the OWL ART command line utilities:
http://art.uniroma2.it/owlart/documentation/utilities.jsf

Hope this helps!

-Osma

> ups, to /-o/ really is /-o output.rdf/.
>
> --
> You received this message because you are subscribed to the Google
> Groups "Skosmos Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to skosmos-user...@googlegroups.com
> <mailto:skosmos-user...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/skosmos-users/43d547cf-9157-4cc7-996d-1c12ce57be56n%40googlegroups.com <https://groups.google.com/d/msgid/skosmos-users/43d547cf-9157-4cc7-996d-1c12ce57be56n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
Osma Suominen
D.Sc. (Tech), Information Systems Specialist
National Library of Finland
P.O. Box 15 (Unioninkatu 36)
00014 HELSINGIN YLIOPISTO
Tel. +358 50 3199529
osma.s...@helsinki.fi
http://www.nationallibrary.fi

[Frederik Elwert]

Hi,

just a side remark: For our own thesaurus, we maintain a python script
that takes a turtle file (coming from VocBench in our case, but that
should be quite generic) and performs a series of modifications required
to make SKOSMOS display a SKOS-XL+SKOS-THES dataset correctly. Maybe
that’s a helpful start for someone having similar issues.

https://github.com/DiGArtefacts/thesaurus/blob/main/normalize.py

(And yes, some parts are rather project specific, and I know there are
other, more generic solutions out there.)

Best,
Frederik



Am 31.10.23 um 10:23 schrieb Osma Suominen:

Prof. Dr. Frederik Elwert

Außerplanmäßiger Professor
Centrum für Religionswissenschaftliche Studien
Ruhr-Universität Bochum

https://dh.ceres.rub.de/