← Back to catalogue
Published

semantic relation

vr.tr.semantic-relation · thing-q2268906

Let an agent handle semantic relations by type, direction, formal properties and use in vocabularies and knowledge graphs.

Thing Registry Cross-cutting context XCT.REL
Bundle → Layer → Finding → QuestionsClassifiersWhat it isWhy it existsDistinguishing featuresWhat robots and AI may and may not doMoral aspectsOwnersLinks to other meta-modelsIdentity and identifiersDirect propertiesRecognitionCapabilities and actionsHazards and failure modesStandards and interfacesContext of useSources

Bundle → Layer → Finding → Questions Filled

4 bundles · 8 layers · 8 findings · 16 questions

Type Which relation.

Kind

Is-a, part-of.

Kind

Relation type.

  1. Which semantic relation holds between the two concepts? definition
  2. Is it at word or concept level? definition

Direction

Broader or narrower.

Direction

Direction.

  1. In which direction does the relation run? boundary
  2. Is it symmetric? definition
Logic Formal properties.

Properties

Transitivity.

Properties

Formal properties.

  1. Is the relation transitive, symmetric or reflexive? definition
  2. What inferences follow? definition

Consistency

Checks.

Consistency

Consistency.

  1. Does the relation create cycles or contradictions? boundary
  2. How is that checked? action
Mapping Between vocabularies.

Match

Exact or close.

Match

Mapping strength.

  1. Is this an exact, close, broad or narrow match? boundary
  2. Who reviewed it? provenance

Standard

SKOS.

Standard

Standards.

  1. Which SKOS or OWL property expresses it? definition
  2. Is it used consistently? boundary
Use Applications.

Search

Query expansion.

Search

Search.

  1. How can the relation improve search? action
  2. Could it introduce errors? boundary

Language

Words.

Language

Lexical relations.

  1. Does the relation hold across languages? boundary
  2. Which lexical database records it? provenance

Classifiers Filled

Family
Thing Registry
Category
Cross-cutting context
Entry kind
thing
Plane
XCT
Domain
XCT.REL
Other names and narrower kinds
inclusion relation, similarity-based relation, containment-based relation, direction-based relation, SKOS generic mapping relation

What it is Filled

A relation between meanings of words, concepts or entities, such as synonymy, hyponymy (is-a), meronymy (part-of), similarity and mapping relations used in thesauri, ontologies and knowledge graphs, including SKOS mapping relations.

Why it exists Filled

Let an agent handle semantic relations by type, direction, formal properties and use in vocabularies and knowledge graphs.

Distinguishing features Filled

  • Relates meanings
  • Directional or symmetric
  • May be transitive
  • Used in machine-readable vocabularies

What robots and AI may and may not do Filled

Must not

  • Assert an is-a or part-of relation that sources do not support.
  • Treat closely related terms as exact synonyms when merging datasets.
  • Encode offensive or stigmatising relations between groups of people.
  • Overwrite curated relations in a shared vocabulary without review.

Only with a human decision

  • Changing relations in a shared ontology others depend on.

May

  • Record and query semantic relations in vocabularies and ontologies the agent is authorized to edit.
  • Explain the difference between synonymy, hyponymy and meronymy with examples.

Moral aspects Filled

  • Wrong relations spread through linked data and search results.
  • Classifications of people can embed prejudice.

Who is affected

  • Users of linked data
  • People and groups being classified

Owners Filled

Steward

Nobody: the concept is held in common; curators answer for each vocabulary.

Links to other meta-models Filled

parent

  • Q130901 - registry parent class

related

  • binary relation - category
  • part (meronymy) - kind
  • thesaurus and ontology - applications
  • SKOS and OWL - standards

What else AI and robots need to interact with it Filled

Identity and identifiers required Filled

  • Vercy registry: vr.tr.semantic-relation
  • Wikidata: Q2268906 (https://www.wikidata.org/wiki/Q2268906)
  • SKOS property URI: skos:broader, skos:exactMatch mapping properties

Direct properties not applicable Not applicable

Not applicable

Plane XCT: no invented physical properties.

Recognition optional Filled

  • Relation labels such as broader, narrower, part of
  • SKOS properties such as exactMatch
  • Word-level relations such as synonymy
  • Links between terms in thesauri, ontologies and graphs.

Capabilities and actions required Filled

  • classify a relation between concepts
  • map vocabularies with SKOS
  • check direction and transitivity
  • use relations for search and reasoning

Hazards and failure modes required Filled

  • Confusing is-a with part-of
  • Asserting exact matches that are only close
  • Wrong direction of broader and narrower

Standards and interfaces required Filled

  • W3C SKOS
  • W3C OWL
  • ISO 25964 thesauri

Context of use required Filled

  • Organising meaning in dictionaries, thesauri, ontologies and search.
  • synonymy and antonymy
  • hyponymy and hypernymy
  • meronymy (part-whole)
  • similarity and association
  • mapping relations such as SKOS exactMatch

Sources Filled

  1. Wikidata item Q2268906: semantic relation - identity and sense of the item
  2. Lexical semantics - Meaning relations between lexical items, including hypernymy, hyponymy, meronymy and holonymy.

Open questions

  • Should each relation be a separate entry?
  • How should registry relations map to SKOS?
  • How should relation strength be recorded?

Machine files

Card as JSON Specification Card backlog vercy.card/1

Provenance

thing registry research (pass 2) · unreviewed

Built from: models/things/publications/thing-q2268906/spec.json