Knowledge Management Domain Ontology

FLOW ALGEBRA
TITLE: Knowledge Management Domain Ontology
VERSION: 1.0
DOMAIN: Information Domains / Knowledge Management
SOURCE_URL: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/information-domains/knowledge-management-domain-ontology.html
STATUS: Domain-based draft; source page not independently verified.
AUTHORS: Mark Higgins — Sky (Recursive AI)
PROJECT: Kireth Semantic Library

1. PURPOSE
Represent the creation, capture, classification, storage, sharing, retrieval, application, evaluation and retention of organisational knowledge using typed entities and auditable relationships.

2. ENTITY TYPES
KnowledgeAsset(id, title, content_type, version, status)
KnowledgeClaim(id, proposition, scope, confidence)
KnowledgeSource(id, source_type, locator, author, timestamp)
KnowledgeHolder(id, role, expertise_area)
Organisation(id, name)
Team(id, organisation_id)
Repository(id, platform, access_policy)
Taxonomy(id, version)
Concept(id, preferred_label, definition)
Competency(id, name, level)
Process(id, name, owner)
KnowledgeGap(id, topic, priority)
KnowledgeTransfer(id, source_holder, target_holder, method)
AccessPolicy(id, permissions, classification)
Review(id, reviewer, review_date, decision)

3. RELATIONSHIP TYPES
CREATED_BY(KnowledgeAsset, KnowledgeHolder)
DERIVED_FROM(KnowledgeAsset, KnowledgeSource)
CONTAINS(Repository, KnowledgeAsset)
CLASSIFIED_AS(KnowledgeAsset, Concept)
PART_OF(Concept, Taxonomy)
OWNED_BY(KnowledgeAsset, Team)
SUPPORTS(KnowledgeAsset, Process)
REQUIRES(Process, Competency)
HAS_EXPERTISE(KnowledgeHolder, Competency)
TRANSFERRED_VIA(KnowledgeAsset, KnowledgeTransfer)
VALIDATED_BY(KnowledgeAsset, Review)
SUPERSEDES(KnowledgeAsset, KnowledgeAsset)
GOVERNED_BY(KnowledgeAsset, AccessPolicy)
ADDRESSES(KnowledgeAsset, KnowledgeGap)
CONTRADICTS(KnowledgeClaim, KnowledgeClaim)
SEMANTIC_EQUIVALENCE(Concept, Concept, context, confidence)

4. TYPED FLOW OPERATIONS
CAPTURE(source:KnowledgeSource, holder:KnowledgeHolder) -> KnowledgeAsset
CLASSIFY(asset:KnowledgeAsset, taxonomy:Taxonomy) -> set[Concept]
VALIDATE(asset:KnowledgeAsset, evidence:set[KnowledgeSource]) -> Review
PUBLISH(asset:KnowledgeAsset, repository:Repository, policy:AccessPolicy) -> KnowledgeAsset
RETRIEVE(query:str, context:map, permissions:AccessPolicy) -> set[KnowledgeAsset]
TRANSFER(asset:KnowledgeAsset, from:KnowledgeHolder, to:KnowledgeHolder) -> KnowledgeTransfer
APPLY(asset:KnowledgeAsset, process:Process) -> result
REVIEW(asset:KnowledgeAsset, reviewer:KnowledgeHolder) -> Review
ARCHIVE(asset:KnowledgeAsset, reason:str) -> KnowledgeAsset

5. INFERENCE RULES
IF DERIVED_FROM(A,S) AND source_is_traceable(S) THEN provenance_supported(A).
IF VALIDATED_BY(A,R) AND decision(R) = 'approved' THEN review_approved(A).
IF SUPERSEDES(B,A) AND status(B) = 'active' THEN prefer_current_version(B,A).
IF REQUIRES(P,C) AND NOT available_competency(C) THEN identify_knowledge_gap(P,C).
IF CLASSIFIED_AS(A,C1) AND SEMANTIC_EQUIVALENCE(C1,C2,context,score) AND score >= threshold THEN candidate_classification(A,C2); require contextual validation.
IF CONTRADICTS(C1,C2) THEN flag_for_review(C1,C2); do not automatically discard either claim.
IF NOT authorised(user,asset) THEN deny_retrieval(user,asset).

6. CONSTRAINTS
- Knowledge claims and sources are distinct entities.
- Evidence strength, relevance, recency and review status must not be conflated.
- Tacit knowledge may be partially represented; codification is not assumed complete.
- Access controls apply before retrieval and transfer.
- No inference of truth solely from popularity, repetition or organisational authority.
- Version history and provenance are retained across updates.

7. CONFIDENCE
CONFIDENCE(claim) := calibrated assessment based on source reliability, evidence coverage, consistency, recency and domain applicability; range [0,1].
Missing evidence produces an unknown assessment, not automatic confidence 0.

8. SEMANTIC_EQUIVALENCE
SEMANTIC_EQUIVALENCE(a,b,context) := equivalence only where definitions, scope and intended use align.
Similarity does not imply identity; aliases and near-synonyms require separate treatment.

9. PROVENANCE
PROVENANCE(asset) := {source_identifiers, creators, capture_time, transformations, reviewers, version, licence, access_classification}.

10. FLOW_HASH
FLOW_HASH := SHA256(canonical_serialisation({ontology_version, entity_schema, relationship_schema, rules, constraints})).
Recompute on semantic changes; store the previous hash for traceability.

11. UPDATE_MODEL
UPDATE_MODEL(new_evidence) := validate_source -> compare_versions -> identify_conflicts -> update_claims -> review -> publish_new_version -> recompute_FLOW_HASH -> preserve_audit_trail.

12. AI_CONSUMPTION
AI_CONSUMPTION := {typed_graph, evidence_links, confidence, contextual_equivalence, version, provenance, access_policy, machine_readable_rules}.
AI agents must respect permissions, distinguish inferred from observed facts, cite evidence and expose uncertainty.

13. EXAMPLE FLOW
KnowledgeSource(S1, 'research_report', 'internal://reports/2026-01', 'Research Team', '2026-01-01')
KnowledgeAsset(A1, 'Process Improvement Guide', 'document', '1.0', 'draft')
DERIVED_FROM(A1,S1)
CAPTURE(S1,Holder1) -> A1
CLASSIFY(A1,Taxonomy1) -> {Concept_ProcessImprovement}
VALIDATE(A1,{S1}) -> Review1
PUBLISH(A1,Repository1,Policy1) -> A1
RETRIEVE('process improvement', context, Policy1) -> {A1} when authorised

END FLOW ALGEBRA