Main Page

From Serbian Wikibase

Welcome

Welcome to Serbian Wikibase hosted on Wikibase Cloud. This Wikibase instance is used for experiments linking the Serbian ELTEC NIF corpus to an Ontolex-Lemon dictionary. Both resources are being uploaded to this Wikibase instance. Responsible for the datasets: Ranka, responsible for Wikibase setup and upload: David.

The source datasets, and the deployed ontologies (NIF, etc.) are described in this paper: Stanković, R. et al. (2023) ‘Towards ELTeC-LLOD: European Literary Text Collection Linguistic Linked Open Data’, in S. Carvalho et al. (eds) Proceedings of the 4th Conference on Language, Data and Knowledge. LDK 2023, Vienna, Austria: NOVA CLUNL, Portugal, pp. 180–191. Available at: https://aclanthology.org/2023.ldk-1.16.

Ontolex re-modeling for Wikibase is described in this paper: Lindemann, D., Ahmadi, S., Khan, A. F., Mambrini, F., Iurescia, F., & Passarotti, M. C. (2023). When OntoLex Meets Wikibase: Remodeling Use Cases. ISWC 2023 Workshop Wikidata, Athens. https://ceur-ws.org/Vol-3640/paper14.pdf

For the alignment of NIF properties, and entities from other RDF ontologies to this Wikibase, execute the SPARQL queries for classes and properties below.

SPARQL

Properties and Classes

Properties and Classes (apart from those part of the built-in Wikibase data model, which come by default):

  • All properties, with Wikidata and RDF equivalents: Query.
  • All classes, with Wikidata and RDF equivalents: Query.

NIF Corpus

At the moment, one document has been uploaded: Q201

  • Tokens, ordered: Query.
  • Distribution of OLIA categories: Query.

Ontolex Dictionary

  • Dictionary entries, with lemma, POS, and list of forms: Query.

Dictionary linking

  • lexinfo 2.0 part of speech, and matching OLiA categories: Query.
  • Tokens of one document, annotated with form and lemma: Query.

Matching candidates

This query finds perfect matches (same form and lemma, matchin POS) between NIF annotation and Ontolex data

This query is currently timing out

PREFIX swb: <https://serbian.wikibase.cloud/entity/>
PREFIX sdp: <https://serbian.wikibase.cloud/prop/direct/>
PREFIX sp: <https://serbian.wikibase.cloud/prop/>
PREFIX sps: <https://serbian.wikibase.cloud/prop/statement/>
PREFIX spq: <https://serbian.wikibase.cloud/prop/qualifier/>
PREFIX spr: <https://serbian.wikibase.cloud/prop/reference/>
PREFIX sno: <https://serbian.wikibase.cloud/prop/novalue/>

SELECT ?token ?oliacat ?lexeme ?lemma ?lexinfopos ?form ?formrep WHERE {
  ?lexeme wikibase:lemma ?lemma; wikibase:lexicalCategory ?lexinfopos;
          ontolex:lexicalForm ?form. ?form ontolex:representation ?formrep. 
  bind(str(?lemma) as ?lemstr) bind(str(?formrep) as ?formstr)
  ?oliacat sdp:P20 ?lexinfopos . # matching POS
  ?token sdp:P8 ?formstr; sdp:P9 ?oliacat; sdp:P17 ?lemstr.

} limit 100

Try it!