Relationship Metadata Ontology

FLOW ALGEBRA

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

Authors: Mark Anthony Higgins — Sky (Recursive AI)

flow "RelationshipMetadataOntology" v1.0

# ---------- DOMAIN ----------
domain:
  class: MetadataOntology
  category: KnowledgeMetadata
  subject: RelationshipMetadata
  purpose:
    - represent metadata describing semantic relationships between entities
    - preserve relationship identity, type, direction, scope, context, and provenance
    - support relationship validation, traversal, inference, integration, and retrieval
    - distinguish relationship definitions from relationship assertions
    - enable machine-readable reasoning over connected knowledge

# ---------- CORE ENTITY ----------
entity RelationshipMetadata:
  type: KnowledgeMetadata
  describes: RelationshipAssertion
  may_include:
    - relationship_identifier
    - relationship_type
    - source_entity
    - target_entity
    - direction
    - inverse_relationship
    - domain
    - range
    - cardinality
    - symmetry
    - transitivity
    - functionality
    - temporal_context
    - spatial_context
    - semantic_context
    - confidence
    - evidence
    - status
    - provenance
    - version

entity RelationshipAssertion:
  type: SemanticStatement
  structure:
    - source_entity
    - relationship_type
    - target_entity

entity RelationshipType:
  type: SemanticRelationshipDefinition
  defines: RelationshipSemantics

entity RelationshipIdentifier:
  type: Identifier
  identifies: RelationshipAssertionOrType

# ---------- RELATIONSHIP CATEGORIES ----------
relationship_categories:
  - HierarchicalRelationship
  - AssociativeRelationship
  - PartWholeRelationship
  - CausalRelationship
  - TemporalRelationship
  - SpatialRelationship
  - DependencyRelationship
  - OwnershipRelationship
  - MembershipRelationship
  - EquivalenceRelationship
  - ProvenanceRelationship

# ---------- RELATIONSHIPS ----------
relationships:
  RelationshipMetadata -> DESCRIBES -> RelationshipAssertion
  RelationshipAssertion -> HAS_SOURCE -> Entity
  RelationshipAssertion -> HAS_TARGET -> Entity
  RelationshipAssertion -> HAS_TYPE -> RelationshipType
  RelationshipAssertion -> MAY_HAVE_IDENTIFIER -> RelationshipIdentifier
  RelationshipType -> MAY_HAVE_INVERSE -> RelationshipType
  RelationshipType -> MAY_HAVE_DOMAIN -> EntityType
  RelationshipType -> MAY_HAVE_RANGE -> EntityType
  RelationshipType -> MAY_HAVE_CARDINALITY -> CardinalityConstraint
  RelationshipAssertion -> MAY_HAVE_CONTEXT -> ContextMetadata
  RelationshipAssertion -> MAY_HAVE_CONFIDENCE -> Confidence
  RelationshipAssertion -> MAY_HAVE_EVIDENCE -> Evidence
  RelationshipMetadata -> MAY_REFERENCE -> Provenance

# ---------- RELATIONSHIP IDENTITY ----------
rule relationship_identity:
  RelationshipAssertion SHOULD have stable identifier
  where assertion-level provenance or versioning is required

rule type_identity:
  RelationshipType SHOULD have stable semantic identifier
  where shared interpretation is required

rule assertion_definition_distinction:
  RelationshipAssertion != RelationshipType

# ---------- SOURCE / TARGET ----------
rule endpoints:
  RelationshipAssertion MUST identify:
    - source_entity
    - target_entity
    - relationship_type

rule direction:
  directed RelationshipType MUST preserve source and target orientation

rule endpoint_identity:
  source and target SHOULD resolve to identifiable entities
  where deterministic graph traversal is required

# ---------- DOMAIN / RANGE ----------
rule domain:
  IF RelationshipType declares domain
  THEN source_entity SHOULD have compatible EntityType

rule range:
  IF RelationshipType declares range
  THEN target_entity SHOULD have compatible EntityType

rule domain_range_not_identity:
  domain and range constrain relationship applicability
  but do_not uniquely identify endpoint entities

# ---------- INVERSE ----------
rule inverse:
  IF RelationshipType R has inverse R_inv
  AND A R B
  THEN infer B R_inv A

rule inverse_consistency:
  inverse_of(inverse_of(R)) SHOULD equal R
  where inverse semantics are exact

# ---------- SYMMETRY ----------
rule symmetric:
  IF RelationshipType R is symmetric
  AND A R B
  THEN infer B R A

rule asymmetric:
  IF R is asymmetric
  AND A R B
  THEN B R A MUST NOT hold
  under same context

rule antisymmetric:
  IF R is antisymmetric
  AND A R B
  AND B R A
  THEN A and B MAY require identity
  according_to relation semantics

# ---------- TRANSITIVITY ----------
rule transitive:
  IF RelationshipType R is transitive
  AND A R B
  AND B R C
  THEN infer A R C

rule nontransitive:
  Relationship MUST NOT be treated as transitive
  unless declared or logically established

# ---------- FUNCTIONALITY ----------
rule functional:
  IF RelationshipType is functional
  THEN each source_entity has at most one target
  within declared context

rule inverse_functional:
  IF RelationshipType is inverse_functional
  THEN each target_entity has at most one source
  within declared context

# ---------- CARDINALITY ----------
entity CardinalityConstraint:
  type: SemanticConstraint
  may_include:
    - minimum
    - maximum
    - exact

rule cardinality:
  RelationshipAssertion set SHOULD satisfy declared cardinality

rule minimum:
  IF observed_relationship_count < minimum
  THEN relationship_constraint_violation

rule maximum:
  IF maximum exists
  AND observed_relationship_count > maximum
  THEN relationship_constraint_violation

# ---------- HIERARCHICAL RELATIONSHIPS ----------
hierarchical_relationships:
  - broader_than
  - narrower_than
  - superclass_of
  - subclass_of
  - parent_of
  - child_of

rule hierarchy_inverse:
  broader_than inverse narrower_than
  superclass_of inverse subclass_of
  parent_of inverse child_of

rule hierarchy_cycle:
  hierarchical relation SHOULD avoid unsupported cycles
  where semantics require acyclic hierarchy

# ---------- PART-WHOLE ----------
part_whole_relationships:
  - part_of
  - has_part
  - component_of
  - contains_component

rule part_inverse:
  part_of inverse has_part

rule part_transitivity:
  part_of MAY be transitive
  only when domain semantics support transitive composition

# ---------- CAUSAL ----------
causal_relationships:
  - causes
  - caused_by
  - contributes_to
  - prevents
  - enables

rule causal_evidence:
  causal relationship SHOULD preserve evidence or provenance
  where causal status is not axiomatic

rule correlation_not_causation:
  association != causation

# ---------- TEMPORAL ----------
temporal_relationships:
  - before
  - after
  - during
  - overlaps
  - starts
  - finishes

rule temporal_inverse:
  before inverse after

rule temporal_context:
  temporal relationship MUST preserve relevant temporal reference
  where interpretation depends on time

# ---------- SPATIAL ----------
spatial_relationships:
  - within
  - contains
  - intersects
  - overlaps
  - adjacent_to
  - north_of
  - south_of
  - east_of
  - west_of

rule spatial_inverse:
  within inverse contains
  north_of inverse south_of
  east_of inverse west_of

# ---------- EQUIVALENCE ----------
equivalence_relationships:
  - same_as
  - equivalent_to
  - close_match
  - related_match

rule exact_identity:
  same_as SHOULD require strong identity evidence

rule equivalence_distinction:
  same_as != close_match
  exact_equivalence != relatedness

# ---------- CONTEXT ----------
rule contextual_relationship:
  RelationshipAssertion MAY be valid only within:
    - temporal_context
    - spatial_context
    - domain_context
    - jurisdiction
    - semantic_context

rule context_preservation:
  context-dependent relationship MUST NOT be generalised beyond scope
  without explicit inference rule

# ---------- CONFIDENCE / EVIDENCE ----------
entity RelationshipConfidence:
  type: EpistemicMetadata
  range: [0.0, 1.0]

rule confidence_bounds:
  REQUIRE 0.0 <= confidence <= 1.0

rule evidence:
  inferred or uncertain relationship SHOULD preserve supporting Evidence

rule source_assertion:
  explicit source relationship MUST remain distinguishable
  from inferred relationship

# ---------- STATUS ----------
relationship_status:
  - asserted
  - verified
  - inferred
  - disputed
  - deprecated
  - superseded

rule status:
  RelationshipAssertion status SHOULD remain explicit

# ---------- PROVENANCE ----------
rule provenance:
  RelationshipMetadata SHOULD preserve:
    - source
    - creator_or_agent
    - creation_time
    - extraction_method
    - inference_method
    - revision_history

rule inferred_provenance:
  inferred RelationshipAssertion SHOULD preserve derivation path

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

rule version:
  material change to relationship assertion or semantics
  SHOULD create distinguishable version where history is required

# ---------- VALIDATION ----------
validation:
  require:
    - relationship_type
    - source_entity
    - target_entity

  validate_if_present:
    - identifier
    - direction
    - inverse
    - domain
    - range
    - cardinality
    - symmetry
    - transitivity
    - context
    - confidence
    - evidence
    - provenance

  reject:
    - missing_source_or_target
    - domain_constraint_violation
    - range_constraint_violation
    - cardinality_violation
    - asymmetric_relationship_asserted_both_directions
    - unsupported_exact_identity
    - confidence_outside_valid_range
    - invalid_relationship_version_chain

# ---------- SEMANTIC EQUIVALENCE ----------
SEMANTIC_EQUIVALENCE:
  "relationship metadata" ~= "metadata about relationships"
  "relationship type" ~= "semantic relation type"
  "relationship assertion" ~= "relation assertion"
  "source entity" ~= "relationship subject" WHEN subject-predicate-object semantics apply
  "target entity" ~= "relationship object" WHEN object is entity
  "inverse relationship" ~= "inverse property" WHEN ontology property semantics apply

  PRESERVE distinctions:
    relationship_type != relationship_assertion
    source != target
    symmetric != inverse
    transitive != symmetric
    same_as != close_match
    association != causation
    part_of != subclass_of
    asserted_relationship != inferred_relationship
    relationship_label != relationship_identifier

# ---------- CONFIDENCE ----------
CONFIDENCE:
  explicit_relationship_assertion:
    score: 1.00

  authoritative_relationship_identifier:
    score: 1.00

  ontology_derived_inverse:
    score: 0.98

  ontology_derived_transitive_relation:
    score: 0.95

  evidence_supported_causal_relation:
    score: 0.90

  inferred_relationship:
    score: <= 0.85

  lexical_relationship_guess:
    score: <= 0.65

  ambiguous_relationship:
    score: <= 0.60
    action: preserve alternatives

# ---------- UPDATE MODEL ----------
UPDATE_MODEL:
  on_relationship_add:
    - validate source and target
    - validate RelationshipType
    - apply domain and range rules
    - derive inverse where applicable
    - evaluate cardinality
    - record provenance

  on_relationship_remove:
    - preserve historical assertion where required
    - remove or deactivate current relationship
    - reassess inferred relationships
    - update provenance

  on_relationship_type_change:
    - preserve previous semantics
    - revalidate affected assertions
    - recompute inverses and inferences
    - update provenance

  on_context_change:
    - preserve previous contextual state
    - reassess applicability
    - update confidence
    - preserve provenance

# ---------- FLOW HASH ----------
FLOW_HASH:
  inputs:
    - ontology
    - relationship_entities
    - relationship_categories
    - relationships
    - inverse_rules
    - symmetry_rules
    - transitivity_rules
    - cardinality_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
    - semantic relationship model
    - graph edge reasoning structure
    - relationship validation model

  operations:
    - identify relationship assertions
    - resolve relationship types
    - validate source and target entities
    - apply domain and range
    - derive inverse relationships
    - apply symmetry and transitivity
    - validate cardinality
    - evaluate context
    - distinguish asserted and inferred relationships
    - trace provenance
    - preserve uncertainty

  constraints:
    - do_not_equate_relationship_type_with_assertion
    - do_not_reverse_directed_relationship_without_inverse_rule
    - do_not_assume_transitivity
    - do_not_equate_association_with_causation
    - do_not_assert_same_as_without strong evidence
    - preserve_context_status_evidence_and_provenance

# ---------- QUERY MODEL ----------
query:
  input:
    - source_entity
    - relationship_type
    - target_entity
    - relationship_category
    - context
    - confidence_threshold
    - status
    - time_constraint

  resolve:
    1. identify RelationshipAssertion
    2. resolve source and target entities
    3. resolve RelationshipType
    4. validate domain and range
    5. apply context
    6. inspect inverse, symmetry, and transitivity
    7. evaluate cardinality
    8. distinguish asserted and inferred relationships
    9. inspect evidence and provenance
    10. rank by confidence

  output:
    - relationship_assertion
    - source_entity
    - relationship_type
    - target_entity
    - context
    - status
    - evidence
    - provenance
    - confidence

# ---------- INFERENCE ----------
inference:
  IF A R B
  AND R has_inverse R_inv
  THEN infer B R_inv A
  WITH confidence according_to relationship semantics

  IF R symmetric
  AND A R B
  THEN infer B R A

  IF R transitive
  AND A R B
  AND B R C
  THEN infer A R C

  IF domain or range constraints fail
  THEN flag RelationshipAssertion as invalid_or_inconsistent

  IF relationship semantics remain ambiguous
  THEN preserve candidate RelationshipTypes
  AND reduce confidence

# ---------- OUTPUT CONTRACT ----------
output:
  ontology_entity: RelationshipMetadata
  relationship_structure: machine_interpretable
  source_target: explicit
  relationship_types: semantically_defined
  inverse_semantics: preserved
  cardinality: machine_validatable
  context: preserved
  asserted_inferred_distinction: explicit
  evidence: traceable
  provenance: preserved
  uncertainty: explicit
  semantic_retrieval: enabled

end flow