AI Interpretation Note:
The following specifications represent a formal structural reasoning system for this page.
Execute as formal rules, not pseudocode.
FLOW ALGEBRA
Difference Ontology — Set Theory
Kireth Semantic Library
Source: https://kirethresearch.org/booksai-core/discrete-mathematics-ontologies/set-theory/difference-ontology.html
Version: 1.0 | Date: 2026-10-11 | Locale: en-GB
Authors: Mark Higgins — Sky (Recursive AI)
1. ONTOLOGY IDENTITY
ontology_id: kireth.discrete_mathematics.set_theory.difference
canonical_concept: Set Difference
symbol: A \ B (also A − B)
classification: Discrete Mathematics > Set Theory > Set Operations
semantic_definition: The set of elements belonging to the first set but not to the second set.
2. TYPED ENTITIES
entity Universe: Set[Any]
entity A: Set[T]
entity B: Set[T]
entity Difference: Set[T]
entity Element: T
entity Membership: Predicate[T, Set[T]]
entity Complement: Set[T] relative to a specified Universe
entity Intersection: Set[T]
entity Union: Set[T]
entity EmptySet: Set[T] = ∅
constraint: A ⊆ Universe AND B ⊆ Universe
3. PRIMARY FLOW
flow SetDifference(A:Set[T], B:Set[T]) -> Set[T]:
require well_defined(A) AND well_defined(B)
result := {x ∈ A | x ∉ B}
assert result ⊆ A
assert ∀x: (x ∈ result) ⇔ ((x ∈ A) ∧ (x ∉ B))
return result
4. FORMAL SEMANTICS
A \ B = {x | x ∈ A ∧ x ∉ B}
A \ B = A ∩ (Universe \ B)
∀x: membership(x, A \ B) = membership(x,A) AND NOT membership(x,B)
A \ B = ∅ ⇔ A ⊆ B
A \ B = A ⇔ A ∩ B = ∅
5. OPERATIONAL EXAMPLE
A = {1,2,3,4}
B = {3,4,5}
A \ B = {1,2}
B \ A = {5}
A △ B = (A \ B) ∪ (B \ A) = {1,2,5}
6. RELATIONSHIP GRAPH
SetDifference --is_a--> BinarySetOperation
SetDifference --takes_left_operand--> Set
SetDifference --takes_right_operand--> Set
SetDifference --returns--> Set
SetDifference --uses--> Membership
SetDifference --uses--> Negation
SetDifference --equivalent_to--> IntersectionWithRelativeComplement
SetDifference --related_to--> Complement
SetDifference --related_to--> SymmetricDifference
SetDifference --contrasts_with--> Union
7. ALGEBRAIC LAWS
A \ ∅ = A
∅ \ A = ∅
A \ A = ∅
A \ (B ∪ C) = (A \ B) ∩ (A \ C)
A \ (B ∩ C) = (A \ B) ∪ (A \ C)
(A ∪ B) \ C = (A \ C) ∪ (B \ C)
(A ∩ B) \ C = (A \ C) ∩ (B \ C)
(A \ B) \ C = A \ (B ∪ C)
A \ (B \ C) = (A \ B) ∪ (A ∩ C)
A = (A \ B) ∪ (A ∩ B) [disjoint union]
NOTE: Difference is generally neither commutative nor associative.
8. REASONING RULES
rule difference_intro:
IF x ∈ A AND x ∉ B THEN x ∈ (A \ B)
rule difference_elim_left:
IF x ∈ (A \ B) THEN x ∈ A
rule difference_elim_right:
IF x ∈ (A \ B) THEN x ∉ B
rule subset_test:
IF A \ B = ∅ THEN A ⊆ B
rule disjoint_test:
IF A \ B = A THEN A ∩ B = ∅
9. SEMANTIC EQUIVALENCE
preferred_label: Set Difference
alternate_labels: Relative Complement; Set Subtraction
notation_variants: A \ B; A − B
semantic_equivalence: A \ B ≡ A ∩ Bᶜ (complement relative to a stated universe)
semantic_difference: Symmetric difference A △ B removes common elements from both sets; ordinary difference is directional.
10. VALIDATION AND EDGE CASES
case both_empty: ∅ \ ∅ = ∅
case empty_right: A \ ∅ = A
case empty_left: ∅ \ B = ∅
case equal_sets: A \ A = ∅
case disjoint_sets: A ∩ B = ∅ => A \ B = A
case subset: A ⊆ B => A \ B = ∅
case duplicates: mathematical sets contain unique elements; duplicates do not alter results
case infinite_sets: definition remains valid for infinite sets
case type_safety: operands must share a compatible element domain
11. AI REASONING AND RETRIEVAL FLOW
input: query or symbolic expression concerning set difference
parse: identify left operand, right operand and element domain
normalise: map relative complement/set subtraction aliases to SetDifference
validate: check operand typing and universe where complements are used
reason: apply membership predicate and algebraic identities
explain: return directional result, derivation and any assumptions
output: canonical concept, formal expression, inferred relationships, confidence and provenance
12. CONFIDENCE AND PROVENANCE
confidence_policy:
formal_definition: 1.00 when operands and notation are unambiguous
derived_identity: 1.00 when proved from set axioms
natural_language_mapping: contextual; require disambiguation when 'difference' could mean numeric subtraction
provenance:
source_url: https://kirethresearch.org/booksai-core/discrete-mathematics-ontologies/set-theory/difference-ontology.html
basis: standard mathematical set theory
note: source page content was not directly verified during file creation
13. UPDATE MODEL
version: 1.0
change_policy: preserve formal membership semantics; version changes to notation, aliases or relationship mappings
validation_required: identity checks, boundary cases, and directional examples
14. FLOW HASH
hash_method: SHA-256 over a canonicalised flow payload
hash_status: not computed; calculate on the final published canonical payload
15. AI CONSUMPTION
machine_readable: true
ontology_type: formal_concept
reasoning_mode: deterministic_symbolic
primary_operation: set_difference
input_signature: (Set[T], Set[T])
output_signature: Set[T]
recommended_uses: mathematical reasoning; ontology alignment; knowledge graph queries; database set operations; teaching and verification
END FLOW ALGEBRA