MarkLogic 12 Product Documentation

entity.skosDictionary

entity.skosDictionary(
   graph as String,
   [lang as String],
   [options as String[]]
) as cts.entityDictionary

Summary

Construct an entity dictionary from a SKOS ontology loaded into the database as triples. The mapping from a SKOS ontology is as follows:

Usage Notes

For best results it is strongly recommended you construct ontologies using skos:ConceptScheme with explicit skos:inScheme properties.

The following list describes the mapping from a SKOS ontology to entity dictionary entries. NOTE: The "skos" prefix here is shorthand for the namespace "http://www.w3.org/2004/02/skos/core#".

• Each skos:Concept becomes an entity.
• Each skos:ConceptScheme provides an entity type.
• The concept's IRI becomes the entity's ID.
• The concept's skos:prefLabel becomes the entity's normalized text.
• The concept's skos:prefLabel plus any additional labels (skos:altLabel, skos:hiddenLabel) become entity text that will be used to match.
• If the concept is in a skos:ConceptScheme then that scheme will provide the entity type; otherwise the graph URI will be used as the entity type.
• The rdfs:label or dc:title of the concept scheme is used as the entity type, if one is available; otherwise, the IRI of the concept scheme is used.
• A concept C is considered in a scheme S if any of the following conditions (expressed as triples) are met:
  • C skos:inScheme S
  • S skos:hasTopConcept C
  • S skos:hasTopConcept C2 AND C2 skos:narrowerTransitive C
  • C skos:topConceptOf S
  • C skos:broaderTransitive C2 AND C2 skos:topConceptOf S

  skos:broaderTransitive will be inferred from skos:broader if needed. skos:narrowerTransitive will be inferred from skos:narrower if needed.

If this function cannot derive an entity dictionary entries from the graph, MarkLogic throws the exception XDMP-EMPTYEDICT.

See Also

• Creating a Dictionary From a SKOS Ontology in the Search Developer's Guide
• Creating an Entity Dictionary in the Search Developer's Guide
• entity.dictionaryInsert

Example

'use strict';
declareUpdate();
const entity = require('/MarkLogic/entity');

entity.dictionaryInsert('/ontology/people',
  entity.skosDictionary('http://example.com/people'));