100% found this document useful (1 vote)
75 views27 pages

Relations: Jorge Cobb The University of Texas at Dallas

Relations are formal ways to specify which elements from two or more sets are related to each other. A relation R from set A to set B is a subset of the Cartesian product A x B. Relations can be binary, ternary, or n-ary depending on the number of sets involved. Functions are special cases of relations that map each element of the domain set to exactly one element of the codomain set. There are many properties relations can have including reflexive, symmetric, antisymmetric, and transitive. The transitive closure of a relation R is the smallest transitive relation containing R.

Uploaded by

shri3415
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
75 views27 pages

Relations: Jorge Cobb The University of Texas at Dallas

Relations are formal ways to specify which elements from two or more sets are related to each other. A relation R from set A to set B is a subset of the Cartesian product A x B. Relations can be binary, ternary, or n-ary depending on the number of sets involved. Functions are special cases of relations that map each element of the domain set to exactly one element of the codomain set. There are many properties relations can have including reflexive, symmetric, antisymmetric, and transitive. The transitive closure of a relation R is the smallest transitive relation containing R.

Uploaded by

shri3415
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 27

Relations

Jorge Cobb
The University of Texas at Dallas

1
What are relations?
 Relations are a formal means to specify
which elements from two or more sets are
related to each other
 Examples
 {students} who take {courses}
 {businesses} and their {telephone numbers}
 {integers} and their {divisors}
 {program variables} and the {subroutines} they
are used in
2
Definition of relations
 Let A and B be two sets. A binary relation R
from A to B is a subset of A × B.
 Note that the order of the two sets matters.
 More generally, let A1, A2, ..., An be n sets.
An n-ary relation R on these sets is a subset
of A1 × A2 × ... × An.
 The sets Ai are known as the domains of the
relation, and n as its degree
 Again, the order of the domains matters.
3
Example
 Let A = {0,1,2} and B = {a,b,c}.
 Then, R = {(0,a), (1,a), (1,b)} is a relation
from A to B.
 For instance,
 (0,a) in R, but (a,0) is not in R (order matters!)
 Note that no element in A is related to c (which is
fine), and 2 is not related to any element in B.
 (0,a) in R is also denoted 0 R a

4
Functions
 A function f : A → B is a special case of a
relation from A to B
 It satisfies the property: for each x  A, there is
exactly one pair (x, f(x)) in the relation
 Relations are more general than functions
 They allow unmapped elements from A
 They allow one-to-many mappings where an
element from A maps to multiple elements from B
 They also allow many-to-one and many-to-many
mappings
 I.e., any subset of A x B is a relation. 5
Relations on a set
 A relation R on a set A is a relation from A to
A, that is, a subset of A × A, i.e. of, A2
 Examples (on the set of integers):
 x is equal to y
 x is greater than y
 x divides y
 Note that in these cases both A and the
relation R are infinite sets

6
Example

7
How many relations are there?
 Relations on a set A with n elements
 A2 has n2 elements. We know that the
number of subsets of a set with m elements
2
is 2m, so the desired number is 2n .
 The number of functions from A to A is
 Is this asymptotically the same? nn
 nn = (2logn)n = 2nlogn which grows slower than
2
2n .
8
Properties of relations
 A relation R on a set A is
 reflexive, if
 x  A, (x, x)  R
 symmetric, if
 xy, (x, y)  R  (y, x)  R
 antisymmetric, if
 xy, (x, y)  R and (y, x)  R  x = y
 transitive, if
 xyz, (x, y)R and (y, z)  R  (x, z)  R
9
 What properties are satisfied by the relation
R = {}?

10
Example
 Consider the “divides” relation on the set of
nonnegative integers: x|y iff x divides y.
 Reflexive?
 NO. 0 does not divide 0. (would be reflexive on
the set of positive integers, though)
 Symmetric?
 NO. 2 divides 4, but not the other way around.
 Antisymmetric? YES
 Transitive? YES 11
Example

 Which of the above relations are reflexive,


symmetric, anti-symmetric and transitive?

12
Example

 Which of the above relations are reflexive,


symmetric, anti-symmetric and transitive?

13
Counting the reflexive
relations
 How many reflexive relations are there on a
set A with n elements?
 Consider the pairs B:{(x, x) with x  A}.
 There are n such pairs, they must occur in the
relation.
 For every pair (x, y) with x ≠ y, (there are n(n-1)
of them) it may occur in the relation or it may not
 Thus, there are 2n(n-1) total reflexive relations.

14
Combining relations
 Since relations are by definition sets, they
can be combined in any way that sets can be
combined, including union, intersection, and
set difference.
 Let R1 = {(x,y) | x < y}
R2 = {(x,y) | x > y}
 What is R1  R2 , R1  R2 , R1  R2 , R2  R1 , and R1  R2

15
Closure
 A closure of a relation is the smallest
superset of the relation that contains the
desired property.
 E.g., the reflexive closure of a binary relation
R on a set X is the smallest reflexive relation
on A that contains R.

16
Closure example
 Give the symmetric closure of the reflexive
closure of the transitive closure of
R = {(2,1),(2,3)}
 Transitive closure :
 {(2,1),(2,3)}
 Reflexive closure :
 {(1,1),(2,1),(2,2),(2,3),(3,3)}
 Symmetric closure:
 {(1,1),(1,2),(2,1),(2,2),(2,3),(3,2),(3,3)}
17
More on closures
 Does every relation have a transitive closure?
 Yes (more on that later)
 Does every relation have a reflexive closure?
 Yes, we saw the formula a few slides ago
 Does every relation have a symmetric
closure?
 Yes, for every pair (a,b) in R add the pair (b,a)
 Does every relation have an anti-symmetric
closure?
 No. In a closure you can only add edges, not take them away. A
relation R has an anti-symmetric closure only if R is anti- 18
symmetric to begin with. (The closure is R itself)
Composition of relations
 The composition of two relations R: A→B and
S: B→C, denoted S○R, is the relation from A
to C containing all pairs (x, z) such that there
is one y  B with (x, y)  R and (y, z)  S.

19
Example
 What is the composite S○R, where
 R = {(1,1),(1,4),(2,3),(3,1),(3,4)}
 S = {(1,0),(2,0),(3,1),(3,2),(4,1)}

20
Composing the parent relation
 Let R be the relation on the set of all people
such that a R b if a is the parent of b.
 If (a,c) in R○R, what is the family relationship
between a and c?
 If (a,d) in R○R○R, what is the family
relationship between a and d?

21
Powers of relations
 The nth power of a relation R for n ≥ 1 is
defined recursively as
 R1 = R
 Rn+1 = Rn ○ R
 Again, let R be the parent relation
 Let (a,x) in Rn, what is the relationship
between a and x?

22
Transitivity and Powers
 Theorem: A relation is transitive if and only if
 Rn  R for all n
 Proof
 We first show that if Rn  R for all n then R is
transitive
 Simple, we must show that if (a,b) and (b,c) are in
R then (a,c) is also in R.
 Note that if (a,b) and (b,c) are in R then (a,c) is in
R2, and R2  R from our assumption, so we are
done.
23
 Continues …
Transitivity and Powers contd..
 We next show that if R is transitive then Rn  R for
all n. We show it by induction on n.
 Base, n = 1, trivial, R1 = R  R
 Assume for all i, i ≤ n, Rn  R, show that
Rn+1  R
 Let (a,c) be any pair in Rn+1, we must show it is in R
 Thus, from composition, there is an element b such that
(a,b) is in R and (b,c) is in Rn
 From ind. hyp., (b,c) is in R (we assumed Rn  R) and
from the previous bullet (a,b) is in R, and since R is
assumed transitive, then (a,c) is also in R, and we are
done. 24
Transitive Closure
 Let R be any relation (transitive or not)
 Let R* be the union of all the powers of R.
 I.e., R* = Un Rn
 Is R* is the transitive closure of R?
 Hmmm, maybe, we shall see.

 E.g., let R be the parent relation


 Let (a,x) in R*, what is the relationship
between a and x? 25
Transitive Closures and
transitivity
 Corollary: A relation R is transitive iff R = R*.
remember that R* = Un Rn
 From the theorem,
 R*  R iff R is transitive,
 and we know that
 R*  R iff R = R*
 The second bullet above is because R* by
definition is a superset of R, so to be a subset
it must be equal to R.
26
Transitive Closure
 Is R* is the transitive closure of R?
 Can you prove it?.
 You have to show that R* is transitive.
 And that every new pair (a,b) introduced that is
not in R is necessary to make R transitive.

 In general, what is Ri? hmmm

27

You might also like