Context Metadata Ontology

FLOW ALGEBRA

Ontology: Concept Metadata Ontology
Domain: Metadata Ontologies > Knowledge Metadata
Canonical URL: https://kirethresearch.org/booksai-core/ontologies/metadata-ontologies/knowledge-metadata/concept-metadata-ontology.html

Authors: Mark Anthony Higgins — Sky (Recursive AI)

flow "ConceptMetadataOntology" v1.0

# ---------- DOMAIN ----------
domain:
  class: MetadataOntology
  category: KnowledgeMetadata
  subject: ConceptMetadata
  purpose:
    - represent metadata describing concepts and conceptual entities
    - preserve concept identity, labels, definitions, scope, relationships, and provenance
    - support semantic disambiguation, ontology alignment, retrieval, and reasoning
    - distinguish concepts from labels, instances, classes, and lexical forms
    - enable machine-readable concept management

# ---------- CORE ENTITY ----------
entity ConceptMetadata:
  type: KnowledgeMetadata
  describes: Concept
  may_include:
    - concept_identifier
    - preferred_label
    - alternative_label
    - definition
    - description
    - scope_note
    - concept_type
    - broader_concept
    - narrower_concept
    - related_concept
    - equivalent_concept
    - ontology_reference
    - vocabulary
    - language
    - status
    - version
    - provenance
    - confidence

entity Concept:
  type: SemanticEntity
  represents: AbstractMeaningOrIdea
  identified_by: ConceptIdentifier

entity ConceptIdentifier:
  type: SemanticIdentifier
  identifies: Concept

entity ConceptLabel:
  type: LexicalRepresentation
  may_be:
    - PreferredLabel
    - AlternativeLabel
    - HiddenLabel
    - HistoricalLabel

entity ConceptDefinition:
  type: SemanticDescription
  defines: Concept

# ---------- RELATIONSHIPS ----------
relationships:
  ConceptMetadata -> DESCRIBES -> Concept
  Concept -> IDENTIFIED_BY -> ConceptIdentifier
  Concept -> HAS_PREFERRED_LABEL -> PreferredLabel
  Concept -> MAY_HAVE_ALTERNATIVE_LABEL -> AlternativeLabel
  Concept -> MAY_HAVE_HIDDEN_LABEL -> HiddenLabel
  Concept -> HAS_DEFINITION -> ConceptDefinition
  Concept -> MAY_HAVE_BROADER -> Concept
  Concept -> MAY_HAVE_NARROWER -> Concept
  Concept -> MAY_BE_RELATED_TO -> Concept
  Concept -> MAY_BE_EQUIVALENT_TO -> Concept
  Concept -> MAY_BELONG_TO -> Vocabulary
  Concept -> MAY_REFERENCE -> Ontology
  ConceptMetadata -> MAY_REFERENCE -> Provenance

# ---------- CONCEPT IDENTITY ----------
rule concept_identity:
  Concept SHOULD have stable ConceptIdentifier
  where persistent semantic reference is required

rule identifier_not_label:
  ConceptIdentifier != ConceptLabel

rule label_not_identity:
  identical labels MUST NOT automatically imply identical concepts

rule concept_not_instance:
  Concept != Instance
  unless ontology explicitly models a concept as an individual

# ---------- LABELS ----------
rule preferred_label:
  Concept SHOULD have at most one PreferredLabel
  per declared language and vocabulary policy

rule alternative_label:
  AlternativeLabel MAY represent:
    - synonym
    - abbreviation
    - acronym
    - spelling_variant
    - common_name

rule hidden_label:
  HiddenLabel MAY support search and matching
  without being preferred for display

rule language:
  lexical labels SHOULD preserve language
  where multilingual interpretation matters

# ---------- DEFINITIONS ----------
rule definition:
  ConceptDefinition SHOULD describe semantic meaning
  sufficiently to distinguish concept from nearby concepts

rule definition_source:
  definition SHOULD preserve source or provenance
  where externally derived

rule definition_not_label:
  ConceptDefinition != ConceptLabel

# ---------- HIERARCHY ----------
rule broader_narrower:
  IF A broader_than B
  THEN B narrower_than A

rule hierarchy_cycle:
  broader/narrower hierarchy SHOULD NOT contain unsupported cycles

rule transitive_broader:
  IF A broader_than B
  AND B broader_than C
  THEN A may_be transitive_broader_of C
  where vocabulary semantics define transitivity

rule hierarchy_not_equivalence:
  broader_or_narrower relationship != semantic equivalence

# ---------- RELATED CONCEPTS ----------
rule related:
  RelatedConcept represents associative semantic relationship

rule related_symmetry:
  IF A related_to B
  THEN B related_to A
  where relation is defined as symmetric

rule related_not_hierarchy:
  related_to MUST NOT automatically imply broader_or_narrower

# ---------- EQUIVALENCE ----------
entity ConceptEquivalence:
  type: SemanticRelationship
  may_be:
    - exact_equivalence
    - close_equivalence
    - contextual_equivalence

rule exact_equivalence:
  exact equivalence SHOULD require matching semantic meaning
  within declared scope

rule close_equivalence:
  close equivalence MUST remain distinguishable from exact equivalence

rule contextual_equivalence:
  contextual equivalence applies only within specified context

# ---------- VOCABULARY / ONTOLOGY ----------
entity Vocabulary:
  type: KnowledgeOrganisationSystem
  contains: Concept

rule vocabulary_context:
  ConceptIdentifier SHOULD preserve vocabulary namespace

rule ontology_alignment:
  Concept MAY map to ontology class, property, individual, or concept
  but mapping type SHOULD be explicit

rule cross_vocabulary_mapping:
  mappings across vocabularies SHOULD preserve:
    - source_concept
    - target_concept
    - mapping_type
    - provenance
    - confidence

# ---------- CONCEPT TYPE ----------
entity ConceptType:
  type: Classification
  examples:
    - category
    - topic
    - phenomenon
    - process
    - property
    - object_type
    - abstract_idea
    - domain_term

rule type_context:
  ConceptType SHOULD be interpreted within vocabulary or ontology context

# ---------- SCOPE ----------
entity ScopeNote:
  type: ContextMetadata
  describes:
    - intended_usage
    - exclusions
    - domain_scope
    - interpretation_constraints

rule scope:
  ScopeNote SHOULD constrain interpretation where concept meaning is context-dependent

# ---------- STATUS ----------
concept_status:
  - active
  - proposed
  - deprecated
  - superseded
  - historical

rule status:
  Concept status SHOULD remain explicit

rule deprecated:
  deprecated Concept SHOULD remain resolvable
  where historical references exist

rule superseded:
  superseded Concept SHOULD reference replacement Concept where available

# ---------- VERSION ----------
entity ConceptVersion:
  type: VersionMetadata
  may_include:
    - version_identifier
    - predecessor
    - successor
    - creation_time
    - change_description

rule semantic_change:
  material change to Concept meaning SHOULD create distinguishable ConceptVersion

rule lexical_change:
  label change alone MAY NOT require new semantic concept identity
  unless vocabulary policy defines otherwise

# ---------- PROVENANCE ----------
rule provenance:
  ConceptMetadata SHOULD preserve available:
    - creator
    - source
    - vocabulary
    - creation_time
    - revision_history
    - mapping_source

rule mapping_provenance:
  concept mappings SHOULD preserve responsible agent or process
  and evidence where available

# ---------- DISAMBIGUATION ----------
operation disambiguate_concept(label, context):
  evaluate:
    - label_match
    - language
    - definition
    - domain
    - vocabulary
    - broader_concepts
    - related_concepts
    - usage_context

  output:
    - candidate_concepts
    - selected_concept_optional
    - confidence

rule ambiguous_label:
  IF one label maps to multiple concepts
  THEN preserve candidates
  AND use context for resolution

# ---------- VALIDATION ----------
validation:
  require:
    - identifiable_concept

  validate_if_present:
    - identifier
    - preferred_label
    - definition
    - language
    - broader_concept
    - narrower_concept
    - related_concept
    - equivalent_concept
    - vocabulary
    - status
    - provenance

  reject:
    - concept_self_broader
    - concept_self_narrower
    - unsupported_hierarchy_cycle
    - contradictory_exact_equivalence
    - duplicate_preferred_label where vocabulary policy forbids it
    - unresolved_identifier_namespace where deterministic interpretation is required

# ---------- SEMANTIC EQUIVALENCE ----------
SEMANTIC_EQUIVALENCE:
  "concept metadata" ~= "metadata about concepts"
  "concept identifier" ~= "semantic concept identifier"
  "preferred term" ~= "preferred label" WHEN vocabulary semantics match
  "synonym" ~= "alternative label" WHEN lexical synonymy is intended
  "broader term" ~= "broader concept" WHEN thesaurus semantics apply
  "narrower term" ~= "narrower concept" WHEN thesaurus semantics apply

  PRESERVE distinctions:
    concept != label
    concept != definition
    concept != instance
    preferred_label != identifier
    alternative_label != equivalent_concept
    related_concept != broader_concept
    close_equivalence != exact_equivalence
    lexical_change != semantic_change

# ---------- CONFIDENCE ----------
CONFIDENCE:
  explicit_concept_identifier:
    score: 1.00

  explicit_definition:
    score: 1.00

  authoritative_exact_mapping:
    score: 0.98

  context_resolved_label:
    score: 0.90

  inferred_hierarchy:
    score: 0.80

  label_only_match:
    score: <= 0.70

  ambiguous_concept:
    score: <= 0.60
    action: preserve candidates

# ---------- UPDATE MODEL ----------
UPDATE_MODEL:
  on_label_change:
    - preserve previous label where historically relevant
    - update lexical metadata
    - preserve concept identity unless semantics changed
    - update provenance

  on_definition_change:
    - assess whether semantic meaning changed
    - create ConceptVersion if material
    - preserve previous definition
    - update provenance

  on_hierarchy_change:
    - preserve previous relationship
    - validate hierarchy cycles
    - update broader/narrower links
    - reassess dependent inference

  on_deprecation:
    - preserve Concept
    - set status = deprecated
    - reference replacement where available
    - preserve historical resolvability

# ---------- FLOW HASH ----------
FLOW_HASH:
  inputs:
    - ontology
    - concept_entities
    - relationships
    - identity_rules
    - label_rules
    - hierarchy_rules
    - equivalence_rules
    - validation
    - semantic_equivalence
    - confidence
    - update_model

  canonicalisation:
    - normalise_whitespace
    - preserve_case_sensitive_identifiers
    - sort_unordered_property_sets
    - preserve_rule_order

  output:
    deterministic_semantic_hash

# ---------- AI CONSUMPTION ----------
AI_CONSUMPTION:
  interpret_as:
    - knowledge metadata ontology
    - concept identity model
    - semantic vocabulary structure
    - concept disambiguation model
    - ontology alignment structure

  operations:
    - identify concepts
    - resolve concept identifiers
    - interpret labels
    - resolve definitions
    - traverse broader/narrower hierarchies
    - inspect related concepts
    - evaluate equivalence
    - align concepts across vocabularies
    - disambiguate labels
    - preserve concept versions
    - trace provenance
    - expose uncertainty

  constraints:
    - do_not_equate_label_with_concept
    - do_not_assume_same_label_means_same_concept
    - do_not_force_exact_equivalence_from lexical_similarity
    - do_not_discard_deprecated_concepts
    - do_not_collapse_hierarchy_into_equivalence
    - preserve_identifier_namespace_context_and_provenance

# ---------- QUERY MODEL ----------
query:
  input:
    - concept_identifier
    - label
    - definition
    - vocabulary
    - ontology
    - broader_concept
    - related_concept
    - language
    - status
    - context

  resolve:
    1. identify candidate Concept
    2. resolve identifier namespace
    3. inspect labels and language
    4. inspect definition
    5. apply vocabulary and ontology context
    6. traverse hierarchy
    7. inspect related and equivalent concepts
    8. apply scope and status
    9. rank candidates by confidence

  output:
    - concept
    - identifier
    - preferred_label
    - definition
    - hierarchy
    - related_concepts
    - mappings
    - status
    - provenance
    - confidence

# ---------- INFERENCE ----------
inference:
  IF authoritative ConceptIdentifier resolves uniquely
  THEN identify Concept
  WITH confidence = 1.00

  IF label, definition, vocabulary, and context uniquely align
  THEN resolve Concept
  WITH confidence >= 0.90

  IF A broader_than B
  THEN infer B narrower_than A
  WITH confidence = 1.00

  IF exact equivalence is explicitly asserted by authoritative vocabulary mapping
  THEN associate concepts
  WITH confidence = 0.98

  IF label remains ambiguous
  THEN return multiple candidate Concepts
  AND preserve uncertainty

# ---------- OUTPUT CONTRACT ----------
output:
  ontology_entity: ConceptMetadata
  concept_identity: machine_interpretable
  identifiers: stable
  labels: language_aware
  definitions: preserved
  hierarchy: traversable
  equivalence: typed
  vocabulary_context: explicit
  versions: traceable
  provenance: preserved
  uncertainty: explicit
  semantic_retrieval: enabled

end flow