Cross-Domain Reasoning Ontology

FLOW ALGEBRA
TITLE: Cross-Domain Reasoning Ontology
VERSION: 1.0
SOURCE_URL: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/ai-semantic-library-domains/cross-domain-reasoning-ontology.html
STATUS: Domain-based draft; source page not independently verified.
AUTHORS: Mark Higgins — Sky (Recursive AI)
PROJECT: Kireth Semantic Library

ONTOLOGY:
  ID: kireth:CrossDomainReasoningOntology
  TYPE: DomainOntology
  PARENT: kireth:AISemanticLibraryDomains
  PURPOSE: Represent and validate reasoning that transfers concepts, constraints, evidence and inferences between distinct knowledge domains.

TYPES:
  Domain: {id: URI, name: string, ontology: URI, version: string}
  Entity: {id: URI, type: URI, domain: URI}
  Concept: {id: URI, definition: string, domain: URI}
  Relation: {id: URI, source_type: URI, target_type: URI, domain: URI}
  Claim: {id: URI, proposition: string, scope: URI, status: enum}
  Evidence: {id: URI, source: URI, method: string, observed_at: datetime?}
  Mapping: {id: URI, source_concept: URI, target_concept: URI, mapping_type: enum, scope: string}
  Constraint: {id: URI, expression: string, scope: URI}
  Inference: {id: URI, premises: list[URI], conclusion: URI, rule: URI}
  ReasoningPath: {id: URI, steps: list[URI], domains: list[URI]}

RELATIONSHIPS:
  belongs_to_domain(Entity|Concept|Claim -> Domain)
  maps_to(Concept -> Concept)
  analogous_to(Concept -> Concept)
  equivalent_under_scope(Concept -> Concept)
  depends_on(Claim -> Claim)
  supported_by(Claim -> Evidence)
  constrained_by(Inference|Mapping -> Constraint)
  derived_from(Inference -> Claim|Evidence)
  contradicts(Claim -> Claim)
  bridges(ReasoningPath -> Domain)

FLOW:
  INPUT: query, source_domains, target_domain, evidence_set, constraints
  1. PARSE(query) -> candidate_entities, concepts, intent
  2. RESOLVE(candidate_entities, source_domains) -> typed_entities
  3. ALIGN(source_domains, target_domain) -> candidate_mappings
  4. VALIDATE_MAPPINGS(candidate_mappings, definitions, scope, constraints)
  5. TRANSFER(valid_mappings, supported_claims) -> candidate_target_claims
  6. CHECK(candidate_target_claims, domain_axioms, counterexamples, provenance)
  7. INFER(verified_premises, valid_rules) -> conclusions
  8. SCORE(conclusions, evidence_quality, mapping_strength, uncertainty)
  9. EMIT(conclusions, reasoning_paths, limitations, provenance)

INFERENCE_RULES:
  R1: IF mapping(A,B) is validated AND claim(A) is supported AND transfer_constraints hold THEN propose claim(B), NOT assert automatically.
  R2: IF two domain concepts share a label BUT differ in definitions THEN reject unqualified equivalence.
  R3: IF source evidence does not cover target scope THEN flag extrapolation and require additional evidence.
  R4: IF a premise is contradicted by stronger evidence THEN suspend dependent conclusions pending review.
  R5: IF all premises are validated AND inference rule is sound within its declared scope THEN derive conclusion with provenance.
  R6: IF a mapping is analogy-only THEN preserve analogy status; do not upgrade to identity.

CONFIDENCE:
  RANGE: [0.0, 1.0]
  FIELDS: evidence_quality, mapping_strength, inference_validity, scope_match, uncertainty
  POLICY: Record calibrated confidence when available; otherwise mark as uncalibrated. Never interpret a heuristic score as an empirical probability.

SEMANTIC_EQUIVALENCE:
  TYPES: exact, contextual, broader, narrower, related, analogous, none
  CONDITIONS: compatible definitions, explicit scope, noncontradictory constraints
  POLICY: Equivalence is not assumed from lexical similarity.

PROVENANCE:
  RECORD: source_uri, source_version, author_or_agent, timestamp, mapping_rule, evidence_ids, transformation_history
  REQUIREMENT: Every cross-domain conclusion retains traceable premises and mapping steps.

FLOW_HASH:
  ALGORITHM: SHA-256
  INPUT: canonical_serialisation(ontology_id, ontology_version, mapping_ids, premise_ids, rule_ids, conclusion_ids)
  OUTPUT: deterministic_digest
  NOTE: Compute digest at execution time; no fabricated hash is supplied.

UPDATE_MODEL:
  TRIGGERS: new_evidence, ontology_revision, mapping_change, contradiction, rule_revision
  ACTIONS: invalidate_affected_paths, revalidate_mappings, recompute_inferences, refresh_confidence, update_provenance, regenerate_flow_hash
  VERSIONING: preserve previous versions and record change history.

AI_CONSUMPTION:
  FORMAT: UTF-8 structured Flow Algebra
  SUPPORTS: typed parsing, ontology linking, evidence-grounded inference, semantic retrieval, explainable reasoning, recursive validation
  SAFETY: distinguish established facts, hypotheses, analogies and speculative extrapolations; do not silently promote any category.

END FLOW ALGEBRA