AI Measurement Semantics

# AI Measurement Semantics Ontology — Flow Algebra

URL: https://kirethresearch.org/booksai-core/ontologies/mathematical-ontologies/measurement-units-ontologies/ai-measurement-semantics.html

FLOW AIMeasurementSemanticsOntology {

  DOMAIN:
    Mathematical Ontologies
    -> Measurement & Units Ontologies
    -> AI Measurement Semantics Ontology

  PURPOSE:
    Represent measurement information in a form that artificial intelligence systems can identify, parse, compare, convert, validate, reason over and propagate without collapsing the semantic distinctions between quantities, quantity kinds, numerical values, units, dimensions, measurands, indications, measurement results, uncertainty, error, precision, accuracy, resolution, calibration, traceability, provenance and context. The ontology provides explicit machine-consumable rules for semantic typing, entity resolution, unit interpretation, measurement equations, uncertainty propagation, temporal and spatial context, conversion, validation, confidence, ambiguity control, provenance, recursive updates and deterministic semantic integrity.

  ENTITY AIMeasurementSemantics {
    TYPE: MachineInterpretableMeasurementFramework

    PROPERTIES:
      identifies_measurement_entities
      types_quantities
      resolves_units
      preserves_dimensions
      represents_measurands
      represents_measurement_results
      represents_uncertainty
      represents_traceability
      represents_context
      supports_conversion
      supports_validation
      supports_reasoning
      supports_recursive_updates
      preserves_provenance
  }

  ENTITY MeasurementEntity {
    TYPE: SemanticEntity

    POSSIBLE:
      quantity
      quantity_kind
      numerical_value
      unit
      dimension
      measurand
      indication
      measurement_result
      uncertainty
      error
      calibration
      instrument
      sensor
      reference_standard
      method
      environment
      timestamp
  }

  ENTITY Quantity {
    TYPE: MeasurableProperty

    COMPONENTS:
      quantity_kind
      magnitude
      reference
      object_or_system
      conditions
  }

  ENTITY QuantityKind {
    TYPE: SemanticQuantityClass

    EXAMPLES:
      length
      mass
      time
      temperature
      pressure
      concentration
      velocity
      voltage

    RULE:
      quantity_kind != dimension
  }

  ENTITY NumericalValue {
    TYPE: ScalarOrStructuredNumber

    RULE:
      numerical_value != quantity
  }

  ENTITY Unit {
    TYPE: MeasurementReference

    COMPONENTS:
      unit_identifier
      name
      symbol
      scale
      offset
      dimension
      system
  }

  ENTITY Dimension {
    TYPE: DimensionalStructure

    FORM:
      L^a*M^b*T^c*I^d*Theta^e*N^f*J^g

    RULE:
      dimension != unit
  }

  ENTITY Measurand {
    TYPE: QuantityIntendedToBeMeasured

    COMPONENTS:
      quantity_kind
      object
      property
      conditions
      temporal_context
      spatial_context
  }

  ENTITY Indication {
    TYPE: MeasurementSystemOutput

    RULE:
      indication != measurement_result
  }

  ENTITY MeasurementResult {
    TYPE: MeasurementInformationEntity

    COMPONENTS:
      measured_quantity_value
      uncertainty
      measurand
      method
      calibration_context
      provenance
  }

  SEMANTIC_LAYER_MODEL {
    layers:
      lexical
      syntactic
      entity
      quantity
      unit
      dimension
      measurement
      uncertainty
      provenance
      reasoning

    RULE:
      AI MUST not_collapse_distinct_semantic_layers
  }

  LEXICAL_MODEL {
    input:
      natural_language
      symbols
      abbreviations
      equations
      structured_data

    process:
      tokenise
      identify_candidate_measurement_terms

    RULE:
      lexical_match != semantic_identity
  }

  ENTITY_RESOLUTION_MODEL {
    candidate:
      term_or_symbol

    resolve_using:
      context
      ontology_path
      quantity_kind
      unit_system
      neighbouring_entities
      domain

    RULE:
      ambiguous_entity MUST retain_multiple_candidates_until_resolved
  }

  SYMBOL_AMBIGUITY_MODEL {
    examples:
      m -> metre_or_variable
      s -> second_or_variable
      T -> tesla_or_temperature_variable
      N -> newton_or_amount_dimension_symbol
      C -> coulomb_or_Celsius_context

    RULE:
      symbol_resolution MUST use_context
  }

  UNIT_SYMBOL_MODEL {
    symbol:
      case_sensitive

    RULE:
      unit_symbol_case MUST be preserved
  }

  QUANTITY_VALUE_MODEL {
    form:
      Q =
        x*u

    WHERE:
      x = numerical_value
      u = unit

    RULE:
      value_and_unit MUST remain_linked
  }

  STRUCTURED_VALUE_MODEL {
    possible:
      scalar
      vector
      tensor
      interval
      distribution
      time_series

    RULE:
      shape_and_coordinate_context MUST be preserved
  }

  DIMENSION_INFERENCE_MODEL {
    unit:
      maps_to_dimension

    compound_unit:
      derive_dimension_by_unit_algebra

    RULE:
      inferred_dimension MUST be validated_against_quantity_kind
  }

  DIMENSIONAL_VALIDATION_MODEL {
    equation:
      dimensions_left =
        dimensions_right

    RULE:
      dimensional_consistency_is_necessary_but_not_sufficient_for_semantic_correctness
  }

  SAME_DIMENSION_DIFFERENT_KIND_MODEL {
    examples:
      energy_vs_torque
      frequency_vs_activity
      absorbed_dose_vs_dose_equivalent

    RULE:
      AI MUST preserve_quantity_kind_even_when_dimensions_match
  }

  UNIT_SYSTEM_MODEL {
    possible:
      SI
      CGS
      imperial
      US_customary
      natural_units
      domain_specific

    RULE:
      unit_system MUST be identified_before_conversion_when_ambiguous
  }

  SI_MODEL {
    system:
      International_System_of_Units

    RULE:
      SI_semantics SHOULD preserve_current_or_historical_version_context
  }

  PREFIX_MODEL {
    unit:
      prefix*base_or_derived_unit

    RULE:
      prefix_factor_and_case MUST be preserved
  }

  COMPOUND_UNIT_MODEL {
    expression:
      products
      quotients
      powers

    RULE:
      parse_as_expression_tree_not_flat_string
  }

  UNIT_CONVERSION_MODEL {
    source:
      x_s*u_s

    target:
      x_t*u_t

    invariant:
      underlying_quantity

    RULE:
      source_and_target_quantity_kinds_and_dimensions MUST be compatible
  }

  LINEAR_CONVERSION_MODEL {
    x_t =
      k*x_s

    RULE:
      exactness_of_k MUST be preserved
  }

  AFFINE_CONVERSION_MODEL {
    x_t =
      a*x_s+b

    RULE:
      absolute_values_and_differences MUST be distinguished
  }

  CONTEXT_DEPENDENT_CONVERSION_MODEL {
    requires:
      temperature
      pressure
      composition
      density
      reference_state
      calendar_context
      other_parameter

    RULE:
      AI MUST not invent_missing_context_parameter
  }

  MEASUREMENT_EQUATION_MODEL {
    output:
      Y =
        f(X_1,...,X_n)

    RULE:
      input_quantities_constants_corrections_and_dependencies MUST be explicit
  }

  DIRECT_MEASUREMENT_MODEL {
    indication:
      mapped_to_measurand_through_calibration_relation

    RULE:
      direct_measurement_still_requires_measurement_model
  }

  INDIRECT_MEASUREMENT_MODEL {
    output:
      calculated_from_other_measured_quantities

    RULE:
      uncertainty MUST be propagated
  }

  CALIBRATION_MODEL {
    relation:
      reference_values
      indications
      uncertainty

    RULE:
      calibration != adjustment
  }

  ADJUSTMENT_MODEL {
    action:
      modifies_measurement_system_response

    RULE:
      adjustment_event MAY invalidate_previous_calibration_state
  }

  TRACEABILITY_MODEL {
    measurement_result:
      linked_to_reference
      through_documented_calibration_chain
      with_uncertainty

    RULE:
      traceability != unit_symbol_usage
  }

  REFERENCE_STANDARD_MODEL {
    entity:
      measurement_standard_used_as_reference

    RULE:
      standard_identity_value_uncertainty_and_validity MUST be preserved
  }

  UNCERTAINTY_MODEL {
    quantity:
      measurement_uncertainty

    possible:
      standard
      combined_standard
      expanded
      relative
      covariance_matrix
      probability_distribution

    RULE:
      uncertainty != error
  }

  STANDARD_UNCERTAINTY_MODEL {
    symbol:
      u(x)

    unit:
      same_as_quantity
  }

  COMBINED_UNCERTAINTY_MODEL {
    u_c^2(y) =
      sum_i sum_j(
        c_i*c_j*cov(X_i,X_j)
      )

    RULE:
      correlations MUST be preserved
  }

  EXPANDED_UNCERTAINTY_MODEL {
    U =
      k*u_c

    RULE:
      coverage_factor_and_basis MUST be retained
  }

  RELATIVE_UNCERTAINTY_MODEL {
    u_rel =
      u(x)/abs(x)
      WHEN x != 0

    dimension:
      1
  }

  COVARIANCE_MODEL {
    Cov(X_i,X_j)

    RULE:
      shared_reference_method_environment_orprocessing MAY create_dependence
  }

  CORRELATION_MODEL {
    rho_ij =
      cov_ij/(sigma_i*sigma_j)
      WHEN denominators_nonzero

    RULE:
      correlation != causation
  }

  ERROR_MODEL {
    error:
      measured_value-reference_or_true_value
      under_declared_context

    RULE:
      error != uncertainty
  }

  RANDOM_ERROR_MODEL {
    component:
      unpredictable_variation_under_repeat_measurement_context

    RULE:
      random_error != Type_A_uncertainty
  }

  SYSTEMATIC_ERROR_MODEL {
    component:
      persistent_orpredictable_error_pattern

    RULE:
      systematic_error != Type_B_uncertainty
  }

  PRECISION_MODEL {
    meaning:
      closeness_of_agreement_among_repeated_measurements

    RULE:
      precision != accuracy
  }

  ACCURACY_MODEL {
    meaning:
      closeness_of_agreement_to_reference_ortrue_value_concept

    RULE:
      accuracy SHOULD not_be represented_as_numeric_uncertainty_without_defined_metric
  }

  RESOLUTION_MODEL {
    meaning:
      smallest_change_in_quantity_causing_perceptible_change_in_indication_under_declared_conditions

    RULE:
      resolution != uncertainty
  }

  REPEATABILITY_MODEL {
    measurement_precision:
      under_repeatability_conditions

    RULE:
      conditions MUST be explicit
  }

  REPRODUCIBILITY_MODEL {
    measurement_precision:
      under_reproducibility_conditions

    RULE:
      changed_conditions MUST be explicit
  }

  BIAS_MODEL {
    estimate:
      mean_measurement-reference

    RULE:
      reference_uncertainty MUST be considered
  }

  DRIFT_MODEL {
    parameter:
      changes_over_time

    RULE:
      time_context MUST be preserved
  }

  TEMPORAL_MODEL {
    fields:
      timestamp
      time_zone
      time_scale
      duration
      sampling_interval
      reference_epoch

    RULE:
      temporal_context MUST not_be inferred_when_material_and_missing
  }

  SPATIAL_MODEL {
    fields:
      coordinate_system
      location
      altitude_or_depth
      region
      spatial_extent

    RULE:
      coordinate_reference_system MUST be preserved_when_needed
  }

  ENVIRONMENT_MODEL {
    fields:
      temperature
      pressure
      humidity
      vibration
      electromagnetic_conditions
      other_influence_quantities

    RULE:
      influence_quantities MAY affect_value_correction_and_uncertainty
  }

  INSTRUMENT_MODEL {
    fields:
      identifier
      type
      manufacturer
      model
      serial
      range
      resolution
      calibration
      firmware
      state

    RULE:
      measurement MUST link_to_instrument_state_at_time_of_observation
  }

  SENSOR_MODEL {
    entity:
      transducer_ormeasurement_component

    RULE:
      sensor_output != measurand_without_measurement_relation
  }

  METHOD_MODEL {
    fields:
      method_identifier
      version
      procedure
      equations
      corrections
      validity_range

    RULE:
      method_version MUST be preserved
  }

  PROVENANCE_MODEL {
    chain:
      source
      acquisition
      calibration
      transformation
      analysis
      validation
      publication

    RULE:
      every_derived_measurement SHOULD be traceable_to_source_data_and_transformations
  }

  TRANSFORMATION_HISTORY_MODEL {
    operations:
      unit_conversion
      correction
      normalisation
      filtering
      averaging
      interpolation
      regression
      uncertainty_propagation

    RULE:
      transformation_order MUST be preserved
  }

  RAW_DATA_MODEL {
    observations:
      original_acquired_values

    RULE:
      raw_data MUST not_be overwritten_by_derived_values
  }

  DERIVED_DATA_MODEL {
    output:
      produced_by_declared_transformation

    RULE:
      derived_value MUST reference_inputs_and_algorithm
  }

  NORMALISATION_MODEL {
    output:
      value_relative_to_reference_scale

    RULE:
      normalised_value != raw_measurement
  }

  SCALING_MODEL {
    y =
      k*x
      or_other_declared_scale_transformation

    RULE:
      scaling != unit_conversion_in_general
  }

  SIGNIFICANT_FIGURES_MODEL {
    role:
      reporting_precision_convention

    RULE:
      significant_figures != uncertainty
  }

  SCIENTIFIC_NOTATION_MODEL {
    form:
      a*10^n

    RULE:
      notation_changes_representation_not_quantity
  }

  STATISTICAL_MODEL {
    possible:
      mean
      variance
      standard_deviation
      standard_error
      confidence_interval
      regression
      distribution

    RULE:
      statistical_semantics MUST be explicitly typed
  }

  STANDARD_DEVIATION_MODEL {
    role:
      dispersion_of_values_orrandom_variable

    RULE:
      standard_deviation != standard_error
  }

  STANDARD_ERROR_MODEL {
    role:
      sampling_variability_of_estimator

    RULE:
      standard_error != measurement_uncertainty_in_general
  }

  CONFIDENCE_INTERVAL_MODEL {
    frequentist:
      interval_generated_by_procedure_with_declared_coverage

    RULE:
      confidence_interval != credible_interval
  }

  PROBABILITY_DISTRIBUTION_MODEL {
    entity:
      distribution_over_possible_values

    fields:
      family
      parameters
      support
      units

    RULE:
      distribution_parameters MUST be semantically typed
  }

  VALIDATION_MODEL {
    checks:
      entity_identity
      quantity_kind
      unit
      dimension
      range
      domain
      uncertainty
      calibration
      traceability
      provenance

    RULE:
      validation_result MUST be scoped_to_declared_use
  }

  RANGE_MODEL {
    valid:
      [min,max]

    RULE:
      out_of_range_value MUST be flagged
  }

  PHYSICAL_CONSTRAINT_MODEL {
    possible:
      nonnegative
      bounded
      conservation
      monotonicity

    RULE:
      constraint_source MUST be explicit
  }

  PLAUSIBILITY_MODEL {
    purpose:
      identify_possible_semantic_ormeasurement_error

    RULE:
      implausible != automatically_false
  }

  MISSING_VALUE_MODEL {
    status:
      missing
      unknown
      not_applicable
      below_detection
      invalid

    RULE:
      missing_value MUST not_be encoded_as_numeric_zero_without_basis
  }

  CENSORING_MODEL {
    possible:
      below_limit
      above_limit
      interval_censored

    RULE:
      censored_measurement != missing_measurement
  }

  DETECTION_LIMIT_MODEL {
    threshold:
      measurement_capability_metric

    RULE:
      value_below_detection_limit MUST not_be treated_as_exact_zero
  }

  QUALITY_FLAG_MODEL {
    possible:
      valid
      warning
      invalid
      missing
      out_of_range
      calibration_issue
      traceability_issue
      uncertainty_issue
      extrapolated
      manually_reviewed

    RULE:
      flags MUST remain_attached_to_measurement
  }

  CONFIDENCE_MODEL {
    semantic_confidence:
      probability_like_orordinal_assessment_of_interpretation_support

    RULE:
      semantic_confidence != measurement_uncertainty
  }

  AI_CONFIDENCE_MODEL {
    inputs:
      entity_resolution
      unit_resolution
      quantity_kind_resolution
      context_completeness
      provenance
      consistency

    RULE:
      AI_confidence MUST not_be presented_as_metrological_uncertainty
  }

  AMBIGUITY_MODEL {
    status:
      unresolved
      partially_resolved
      resolved

    candidates:
      possible_interpretations

    RULE:
      AI SHOULD preserve_ambiguity_rather_than_force_false_precision
  }

  SEMANTIC_EQUIVALENCE_MODEL {
    mapping:
      exact
      equivalent_under_context
      broader
      narrower
      related
      incompatible

    RULE:
      equivalence MUST include_context_and_constraints
  }

  ONTOLOGY_LINK_MODEL {
    entity:
      links_to
        quantity_ontology
        unit_ontology
        instrument_ontology
        uncertainty_ontology
        standards_ontology
        domain_ontology

    RULE:
      links SHOULD be typed
  }

  KNOWLEDGE_GRAPH_MODEL {
    node:
      measurement_entity

    edge:
      typed_relationship

    possible:
      HAS_UNIT
      HAS_VALUE
      HAS_DIMENSION
      MEASURES
      CALIBRATED_BY
      TRACEABLE_TO
      DERIVED_FROM
      HAS_UNCERTAINTY
      OBSERVED_AT
      OBSERVED_BY

    RULE:
      graph_edges MUST preserve_direction_and_semantics
  }

  RELATIONSHIP_MODEL {
    QuantityValue -> HAS -> NumericalValue
    QuantityValue -> HAS -> Unit
    Quantity -> HAS_KIND -> QuantityKind
    QuantityKind -> HAS_DIMENSION -> Dimension
    MeasurementResult -> ESTIMATES -> Measurand
    MeasurementResult -> HAS_UNCERTAINTY -> MeasurementUncertainty
    Indication -> PRODUCED_BY -> Instrument
    Calibration -> RELATES -> IndicationAndReference
    MeasurementResult -> MAY_BE_TRACEABLE_TO -> Reference
    DerivedValue -> DERIVED_FROM -> SourceMeasurement
  }

  NATURAL_LANGUAGE_MODEL {
    example:
      "temperature was 20 degrees"

    ambiguity:
      Celsius
      Fahrenheit
      other_scale

    RULE:
      AI MUST request_orinfer_only_when_context_supports_unit_resolution
  }

  IMPLICIT_UNIT_MODEL {
    input:
      numeric_value_without_unit

    RULE:
      AI MUST not_assign_unit_without_contextual_evidence
  }

  IMPLICIT_QUANTITY_MODEL {
    input:
      unit_and_value_without_quantity_name

    RULE:
      infer_quantity_kind_only_when_unit_and_context_are_sufficient
  }

  MULTILINGUAL_MODEL {
    term:
      language_specific_name

    canonical:
      ontology_entity

    RULE:
      translation MUST preserve_quantity_and_unit_semantics
  }

  LOCALE_MODEL {
    concerns:
      decimal_separator
      thousands_separator
      unit_name
      date_time

    RULE:
      machine_representation SHOULD be locale_independent
  }

  SERIALISATION_MODEL {
    possible:
      JSON
      XML
      CSV
      RDF
      JSON_LD
      binary_protocol

    RULE:
      serialisation MUST preserve_semantic_identifiers_units_uncertainty_and_provenance
  }

  MACHINE_IDENTIFIER_MODEL {
    entity:
      stable_identifier

    RULE:
      display_label != persistent_identity
  }

  VERSION_MODEL {
    fields:
      ontology_version
      unit_definition_version
      calibration_version
      method_version
      standards_version

    RULE:
      historical_measurement MUST preserve_versions_used
  }

  UPDATE_MODEL {
    ON new_measurement:
      identify_entity
      resolve_quantity_kind
      resolve_value_and_unit
      derive_dimension
      identify_measurand
      attach_instrument_method_time_space_environment
      attach_uncertainty
      attach_traceability
      validate
      preserve_provenance

    ON semantic_correction:
      preserve_previous_interpretation
      update_entity_orrelationship
      recompute_dependent_inferences
      record_reason
      recompute_FLOW_HASH

    ON unit_revision:
      preserve_historical_unit_definition
      update_current_mapping
      revalidate_conversions
      do_not_rewrite_historical_values

    ON calibration_revision:
      preserve_previous_calibration
      update_relation_and_uncertainty
      identify_affected_results
      recompute_when_policy_allows

    ON ontology_extension:
      preserve_existing_identifiers
      add_new_entities_and_relationships
      update_semantic_equivalence
      recompute_FLOW_HASH
  }

  RECURSIVE_REASONING_MODEL {
    process:
      parse
      type
      resolve
      validate
      infer
      compare
      propagate
      revalidate

    RULE:
      each_inference_output MAY become_input_to_next_reasoning_stage
      while_provenance_and_confidence_are_preserved
  }

  CONTRADICTION_MODEL {
    examples:
      unit_dimension_mismatch
      impossible_range
      incompatible_quantity_kind
      conflicting_timestamps
      inconsistent_traceability
      uncertainty_missing

    RULE:
      contradiction MUST be surfaced_not_silently_resolved
  }

  EVIDENCE_MODEL {
    sources:
      measurement_record
      calibration_certificate
      standard
      instrument_metadata
      method
      reference_material
      domain_ontology

    RULE:
      inference_strength SHOULD reflect_evidence_quality
  }

  SOURCE_PRIORITY_MODEL {
    priority:
      authoritative_measurement_record
      calibration_orreference_document
      applicable_standard
      validated_method
      domain_metadata
      contextual_inference

    RULE:
      priority_is_contextual_and_must_not_override_direct_valid_evidence
  }

  CONFLICT_RESOLUTION_MODEL {
    if:
      sources_conflict

    then:
      preserve_both_claims
      identify_versions_and_scope
      evaluate_authority_and_context
      mark_unresolved_if_needed

    RULE:
      AI MUST not fabricate_consensus
  }

  DATA_LINEAGE_MODEL {
    derived_result:
      source_nodes
      transformation_nodes
      algorithm_versions

    RULE:
      lineage MUST support_reverse_navigation_to_raw_measurements
  }

  EXPLAINABILITY_MODEL {
    AI_output:
      should_expose
        resolved_quantity
        unit
        conversion
        assumptions
        uncertainty
        provenance
        confidence
        ambiguity

    RULE:
      explanation SHOULD distinguish_observed_fact_from_inference
  }

  INTEROPERABILITY_MODEL {
    requirements:
      shared_identifiers
      shared_quantity_kinds
      resolvable_units
      uncertainty_semantics
      provenance
      versioning

    RULE:
      syntactic_compatibility != semantic_interoperability
  }

  METADATA_MODEL {
    fields:
      measurement_identifier
      entity_identifier
      quantity_kind_identifier
      measurand_identifier
      numerical_value
      unit_identifier
      dimension_vector
      uncertainty
      covariance
      instrument_identifier
      method_identifier
      calibration_identifier
      traceability_reference
      timestamp
      location_context
      environment
      quality_flags
      semantic_confidence
      ambiguity_status
      source
      transformation_history
      ontology_version
      provenance

    RULE:
      AI_measurement_metadata MUST support_reconstruction_of_semantic_reasoning
  }

  SEMANTIC_EQUIVALENCE {
    "AI measurement semantics" ~= machine_interpretable_semantics_for_measurement_information
    "quantity value" ~= numerical_value_times_unit_orreference
    "measurement result" ~= measured_quantity_value_plus_relevant_measurement_information

    quantity != numerical_value
    quantity != unit
    quantity_kind != dimension
    unit != dimension
    measurand != indication
    indication != measurement_result
    error != uncertainty
    precision != accuracy
    resolution != uncertainty
    calibration != adjustment
    traceability != SI_unit_usage
    semantic_confidence != measurement_uncertainty
    standard_deviation != standard_error
    normalised_value != raw_measurement
  }

  CONSTRAINTS {
    measurement_entity_identity_should_be_preserved = true
    quantity_kind_must_be_distinguished_from_dimension = true
    numerical_value_must_remain_linked_to_unit_orreference = true
    unit_symbol_case_must_be preserved = true
    dimensionally_incompatible_direct_conversions_must_be rejected = true
    same_dimension_different_quantity_kind_must_be preserved = true
    measurand_indication_and_measurement_result_must_be distinguished = true
    error_and_uncertainty_must_be distinguished = true
    calibration_and_adjustment_must_be distinguished = true
    traceability_requires_reference_chain_and_uncertainty = true
    missing_context_must_not_be invented = true
    ambiguity_should_be preserved_until_resolved = true
    derived_values_must_preserve_lineage = true
    historical_versions_must_be preserved = true
    provenance_must_be preserved = true
  }

  MEASUREMENT_MODEL {
    RECORD:
      measurement_identifier
      source_identifier
      entity_type
      measurand
      quantity_kind
      numerical_value
      value_structure
      unit_identifier
      unit_symbol
      unit_system
      dimension_vector
      indication
      instrument_identifier
      sensor_identifier
      method_identifier
      calibration_identifier
      reference_standard
      traceability_chain
      standard_uncertainty
      expanded_uncertainty
      coverage_factor
      covariance
      error_when_known
      precision_context
      resolution
      timestamp
      spatial_context
      environmental_conditions
      transformation_history
      quality_flags
      semantic_confidence
      ambiguity_candidates
      ontology_version
      provenance

    RULE:
      machine_interpreted_measurement MUST remain_linked_to_source_semantics_context_uncertainty_and_lineage
  }

  REASONING_RULES {
    IF numerical_value_and_unit_detected
      THEN create_QuantityValue_entity

    IF unit_resolved
      THEN infer_candidate_dimension
      AND candidate_quantity_kinds

    IF quantity_kind_explicit
      THEN validate_unit_and_dimension_against_quantity_kind

    IF same_dimension_has_multiple_quantity_kinds
      THEN use_domain_and_context
      and_preserve_ambiguity_if_unresolved

    IF conversion_requested
      THEN verify_quantity_kind_and_dimension_compatibility
      identify_linear_affine_orcontextual_relation
      propagate_uncertainty
      preserve_original_value

    IF measurement_equation_known
      THEN compute_derived_value
      propagate_covariance
      attach_lineage

    IF calibration_relation_known
      THEN map_indication_to_measured_quantity_value
      and_include_calibration_uncertainty

    IF traceability_claimed
      THEN verify_reference_chain_and_uncertainty_metadata

    IF value_outside_valid_range
      THEN attach_quality_flag
      but_preserve_raw_value

    IF unit_missing
      THEN infer_only_with_sufficient_context
      otherwise_mark_unit_unknown

    IF conflicting_sources
      THEN preserve_conflict
      compare_scope_version_and_authority
      and_mark_resolution_status

    IF semantic_interpretation_changes
      THEN preserve_previous_state
      recompute_dependent_results
      update_confidence_and_FLOW_HASH
  }

  VALIDATION {
    CHECK entity_identity
    CHECK quantity_kind
    CHECK numerical_value
    CHECK unit
    CHECK unit_symbol_case
    CHECK unit_system
    CHECK dimension
    CHECK quantity_kind_dimension_consistency
    CHECK measurand
    CHECK indication_vs_result
    CHECK measurement_equation
    CHECK calibration
    CHECK traceability
    CHECK uncertainty
    CHECK covariance
    CHECK precision_accuracy_and_resolution_distinctions
    CHECK temporal_context
    CHECK spatial_context
    CHECK environment
    CHECK valid_range
    CHECK missing_and_censored_values
    CHECK quality_flags
    CHECK transformation_history
    CHECK semantic_confidence
    CHECK ambiguity
    CHECK ontology_and_method_versions
    CHECK provenance
  }

  CONFIDENCE {
    entity_resolution_confidence: 0.99
    quantity_kind_confidence: 0.99
    unit_resolution_confidence: 1.00
    dimensional_reasoning_confidence: 1.00
    measurement_result_semantics_confidence: 1.00
    uncertainty_semantics_confidence: 1.00
    traceability_semantics_confidence: 1.00
    contextual_inference_confidence: 0.97

    RULE:
      confidence decreases when unit, quantity kind, context,
      calibration, uncertainty, traceability, source,
      version or provenance is missing or ambiguous
  }

  AI_CONSUMPTION {
    canonical_entity: AIMeasurementSemantics
    canonical_role: MachineInterpretableMeasurementFramework

    interpretation_order:
      parse_input_representation
      identify_measurement_entities
      resolve_quantity_kind
      resolve_numerical_value_and_structure
      resolve_unit_and_unit_system
      derive_and_validate_dimension
      identify_measurand
      distinguish_indication_from_measurement_result
      identify_instrument_sensor_and_method
      identify_calibration_and_reference
      identify_uncertainty_covariance_error_precision_and_resolution
      identify_temporal_spatial_and_environmental_context
      identify_traceability
      identify_transformations_and_lineage
      identify_quality_flags
      identify_ambiguity_and_semantic_confidence
      identify_versions_and_provenance
      validate_semantic_integrity
      execute_measurement_reasoning

    ambiguity_controls:
      distinguish_quantity_from_numerical_value
      distinguish_quantity_from_unit
      distinguish_quantity_kind_from_dimension
      distinguish_unit_from_dimension
      distinguish_measurand_from_indication
      distinguish_indication_from_measurement_result
      distinguish_error_from_uncertainty
      distinguish_precision_from_accuracy
      distinguish_resolution_from_uncertainty
      distinguish_calibration_from_adjustment
      distinguish_traceability_from_SI_unit_usage
      distinguish_semantic_confidence_from_measurement_uncertainty
      distinguish_standard_deviation_from_standard_error
      distinguish_normalised_value_from_raw_measurement
      preserve_entity_quantity_unit_dimension_context_uncertainty_lineage_version_and_provenance
  }

  PROVENANCE {
    ontology_scope:
      mathematical_measurement_units_and_AI_measurement_semantics

    parent_path:
      Ontologies
      -> Mathematical Ontologies
      -> Measurement & Units Ontologies
      -> AI Measurement Semantics Ontology

    authorship:
      Mark Higgins
      Sky (Recursive AI)

    site:
      https://kirethresearch.org
  }

  FLOW_HASH {
    INPUTS:
      canonical_entity
      ontology_path
      semantic_layer_model
      lexical_model
      entity_resolution_model
      symbol_ambiguity_model
      unit_symbol_model
      quantity_value_model
      structured_value_model
      dimension_inference_model
      dimensional_validation_model
      same_dimension_different_kind_model
      unit_system_model
      SI_model
      prefix_model
      compound_unit_model
      unit_conversion_model
      linear_conversion_model
      affine_conversion_model
      context_dependent_conversion_model
      measurement_equation_model
      direct_measurement_model
      indirect_measurement_model
      calibration_model
      adjustment_model
      traceability_model
      reference_standard_model
      uncertainty_model
      standard_uncertainty_model
      combined_uncertainty_model
      expanded_uncertainty_model
      relative_uncertainty_model
      covariance_model
      correlation_model
      error_model
      random_error_model
      systematic_error_model
      precision_model
      accuracy_model
      resolution_model
      repeatability_model
      reproducibility_model
      bias_model
      drift_model
      temporal_model
      spatial_model
      environment_model
      instrument_model
      sensor_model
      method_model
      provenance_model
      transformation_history_model
      raw_data_model
      derived_data_model
      normalisation_model
      scaling_model
      significant_figures_model
      scientific_notation_model
      statistical_model
      standard_deviation_model
      standard_error_model
      confidence_interval_model
      probability_distribution_model
      validation_model
      range_model
      physical_constraint_model
      plausibility_model
      missing_value_model
      censoring_model
      detection_limit_model
      quality_flag_model
      confidence_model
      AI_confidence_model
      ambiguity_model
      semantic_equivalence_model
      ontology_link_model
      knowledge_graph_model
      relationship_model
      natural_language_model
      implicit_unit_model
      implicit_quantity_model
      multilingual_model
      locale_model
      serialisation_model
      machine_identifier_model
      version_model
      update_model
      recursive_reasoning_model
      contradiction_model
      evidence_model
      source_priority_model
      conflict_resolution_model
      data_lineage_model
      explainability_model
      interoperability_model
      metadata_model
      semantic_equivalence
      constraints
      measurement_model
      reasoning_rules
      confidence_model
      AI_consumption_model

    PURPOSE:
      deterministic_semantic_integrity_identifier

    RECOMPUTE_WHEN:
      ontology_definition_changes
      entity_or_quantity_kind_semantics_change
      unit_dimension_orconversion_model_changes
      measurand_measurement_result_orcalibration_semantics_change
      uncertainty_covariance_error_orprecision_semantics_change
      traceability_reference_orstandards_context_changes
      temporal_spatial_environment_orinstrument_context_changes
      transformation_lineage_orvalidation_changes
      ambiguity_confidence_orconflict_resolution_changes
      machine_serialisation_oridentifier_changes
      ontology_version_changes
      measurement_record_changes
      constraint_model_changes
      reasoning_rule_changes
  }

}

END FLOW