Computer Science and Information Engineering

National Chi Nan University

Combinatorial Mathematics
Dr. Justie Su-Tzu Juan

Chapter 7 Relations: The Second

Time Around
§ 7.1 Relations Revisited: Properties of
Relations
Slides for a Course Based on the Text
Discrete & Combinatorial Mathematics (5th Edition)
by Ralph P. Grimaldi

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.1 Relations Revisited: Properties of Relations

Def 7.1 : A, B: sets, R  A  B: R is called a relation from A to B;

R  A  A: R is called a relation on A;
Ex 7.1 :
a) Defined R on Z by a R b or (a, b)  R, if a  b:
R is the ordinary “less than or equal to” relation on Z.
(Z 可改成 Q, R, but not on C)
b) Let n  Z+, Define R on Z by x R y, if n(x – y):
R is the modulo n relation on Z.
ex. n = 7: 9R 2, –3R 11, (14, 0)  R, 3R 7 (3 is not related to 7).
c) Let U = {1, 2, 3, 4, 5, 6, 7}, C  U, C = {1, 2, 3, 6}
Define R on P(U) by A R B, if A  C = B  C
ex: {1, 2, 4, 5} and {1, 2, 5, 7} are related,
X = {4, 5} and Y = {7} are related;
S = {1, 2, 3, 4, 5} and T= {1, 2, 3, 6, 7} are not related (S R T)
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.1 Relations Revisited: Properties of Relations

Def 7.2 : A relation R on A is called reflexive 

 x  A, (x, x)  R.

Ex 7.5 : If A= n, A  A= n2, there are 2n2 relations on A.

How many of these are reflexive?
2(n2–n)
Def 7.3 : A relation R on A is called symmetric
  x, y  A, (x, y)  R  (y, x)  R .

Note : Let A= n

1) How many relations on A are symmetric? 2(n2+n)/2
2) Both reflexive and symmetric? 2(n2–n)/2
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.1 Relations Revisited: Properties of Relations

Def 7.4 : A relation R on A is called transitive

  x, y, z  A, (x, y), (y, z)  R  (x, z)  R.

Def 7.5 : A relation R on A is called antisymmetric

  a, b  A, (a R b and b R a)  a = b.

Note : How many relations of A are antisymmetric? (|A| = n) ?

(2n)(3(n2–n)/2)
1) (x, x)  A  A (n : 2n)
2) (x, y)  A  A, x  y: (a) (y, x)  R , (b) (x, y)  R ,
(c) (x, y), (y, x)  R
( (n2–n)/2 : 3(n2–n)/2)

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.1 Relations Revisited: Properties of Relations

Def 7.6 : A relation R on A is called a partial order or a partial

ordering relation, if R is reflexive, antisymmetric, and
transitive.

Ex 7.14 : EX 7.1(a) is a partial order,

EX 7.1(b) is not (∵ not antisymmetric),

Def 7.7 : An equivalence relation R on a set A is a relation that

is reflexive, symmetric, and transitive.

Ex 7.16 (1/2):
a) Ex 7.1(b) is equivalence relation.

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.1 Relations Revisited: Properties of Relations

b) Let A = {1, 2, 3}, then

R1 = {(1, 1), (2, 2), (3, 3)}
R2 = {(1, 1), (2, 2), (2, 3), (3, 2), (3, 3)}
R3 = {(1, 1), (1, 3), (2, 2), (3, 1), (3, 3)}
R4 = {(1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2, 3), (3, 1), (3, 2), (3, 3)}
are all equivalence relations on A?
c) For a given finite set A = {a1, a2, …, an},
A  A : the largest equivalence relation on A.
R = {(ai, ai) | 1  i  n} : the smallest equivalence relation on A.
(equality relation)
e) If R is a relation on A, then R is both an equivalent relation
and a partial order on A if and only if R is the equality
relation on A.
(c) Spring 2021, Justie Su-Tzu Juan
 Computer Science and Information Engineering
National Chi Nan University

Discrete Mathematics
Dr. Justie Su-Tzu Juan

Chapter 7 Relations: The Second

Time Around
§ 7.2 Computer Recognition: Zero-One
Matrices and Directed Graphs (1)
Slides for a Course Based on the Text
Discrete & Combinatorial Mathematics (5th Edition)
by Ralph P. Grimaldi

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Def 7.8 : A, B, C: sets, R1  A  B, R2  B  C. The
composite relation R1。R2  A  C defined by
R1。R2 = {(x, z) | x  A, z  C, and  y  B with (x, y)
 R1, (y, z)  R2}.

Ex 7.17: A = {1, 2, 3, 4}, B = {w, x, y, z}, C = {5, 6, 7} back

R1 = {(1, x), (2, x), (3, y), (3, z)}  A  B
R2 = {(w, 5), (x, 6)}  B  C
R3 = {(w, 5), (w, 6)}  B  C
R1。R2 ={(1, 6), (2, 6)} R1。R3 = 

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Ex 7.18 : A: employees, B: programming languages,
C = {p1, p2, …, p8}: projects.
R1  A  B : (x, y)  R1 means x is proficient in y,
R2  B  C : (y, z)  R2 means z need y.
 R1。R2 has been used to set up a matching
process between employees and projects on the
basis of employee knowledge of specific
programming languages.

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Thm 7.1 : A, B, C, D : sets, R1  A  B, R2  B  C, R3  C  D.
The R1。(R2。R3) = (R1。R2)。R3.
Proof.
1. R1。(R2。R3)  A  D, (R1。R2)。R3  A  D.
2.  (a, d)  R1。(R2。R3)
  b  B s.t. (a, b)  R1  (b, d)  R2。R3
  c  C s.t. (b, c)  R2  (c, d)  R3
∵ (a, b)  R1  (b, c)  R2  (a, c)  R1。R2
∵ (a, c)  R1。R2  (c, d)  R3
 (a, d)  (R1。R2)。R3
∴ R1。(R2。R3)  (R1。R2)。R3
Similar, (R1。R2)。R3  R1。(R2。R3)
R1。(R2。R3) = (R1。R2)。R3
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Def 7.9 : A: sets, R  A  A. The power of R defined recursively:
(a) R1 = R;
(b) Rn+1 = R。Rn ,  n   +.

Ex 7.19: A = {1, 2, 3, 4}, R = {(1, 2), (1, 3), (2, 4), (3, 2)} back
 R2 = {(1, 4), (1, 2), (3, 4)}
 R3 = {(1, 4)}
 Rn = ,  n  4.

Def 7.10: 1) An mn zero-one matrix E = (eij)mn, (0, 1)-matrix:

 m rows, n columns, each entry is 0 or 1.
2) eij  the entry in the ith row and the jth column of E,
 1  i  m and 1  j  n.

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
1 0 0 1
Ex 7.20 : E  0 1 0 1 is a 3  4 (0, 1)-matrix.
 
1 0 0 0
1) e11 = 1 2) e23 = 0 3) e31 = 1

Note : Use the standard operations of matrix addition and

multiplication with the stipulation that 1 + 1 = 1
(Boolean addition).

(c) Spring 2021, Justie Su-Tzu Juan

 R1 = {(1, x), (2, x), (3, y), (3, z)}  A  B
§ 7.2 Computer Recognition: Zero-One Matrices and
R2 = {(w, 5), (x, 6)}  B  C
Directed Graphs
Ex 7.21: The relation matrices for R1, R2 of Ex 7.17: see
(w) (x) (y) (z) B (5) (6) (7)
(1) 0 1 0 0 ( w) 1 0 0
( 2) 0 1 0 0 ( x) 0 1 0
M(R1) = M(R2) = 
(3) 0 0 1 1 ( y) 0 0 0
   
( 4)  0 0 0 0 ( z ) 0 0 0

A
0 1 0 0 1 0 0 0 1 0
0 1 0 0
0 1 0 0  1 0
M(R1)M(R2) =  = = M(R1。R2)
0 0 1 1 0 0 0 0 0 0
     
0 0 0 0 0 0 0 0 0 0

Note : M(R1)M(R2) = M(R1。R2)

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Ex 7.22 : A = {1, 2, 3, 4}, R = {(1, 2), (1, 3), (2, 4), (3, 2)},
as in Ex 7.19. see
Define the relation matrix for R: M(R) is the 44 (0, 1)-matrix
whose entries mij, for 1  i, j  4, are given by mij = 1, if (i, j)  R,
0, otherwise.
0 1 1 0 0 1 0 1
0 0 0 0 0
0 0 1
M(R) =  (M(R))2 =  = M(R2)
0 1 0 0 0 0 0 1
   
0 0 0 0 0 0 0 0

0 0 0 0
0 0 0 0
(M(R))4 =   R4=
0 0 0 0
 
0 0 0 0
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
In general : A: set, |A| = n, R  A  A, M(R) is the relation
matrix for R :
(a) M(R) = 0 (all 0’s) iff R = 
(b) M(R) = 1 (all 1’s) iff R = A  A
(c) M(Rm) = [M(R)]m, for m +.

Def 7.11 : E = (eij)mn, F = (fij)mn : 2 mn (0, 1)-matrices.

E precedes (or is less than) F, E  F,
 eij  fij,  1  i  m, 1  j  n.
1 0 1 1 0 1
Ex 7.23 :E    , F   EF
0 0 1 0 1 1
  8 (0, 1)-matrices G for which E  G.
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Def 7.12 : For n +, In = (ij)nn is the nn (0, 1)-matrix, where
ij = 1, if i = j;
0, if i  j.

Def 7.13 : Let A = (aij)mn. The transpose of A, Atr = (a*ji)nm,

where a*ji = aij, for all 1  j  n, 1  i  m.
0 1 
   0 0 1
Ex 7.24 : A  0 0 , A  
tr

1 1  1 0 1

Def : 1) 00 = 01 = 10 = 0, 11 = 1 (usual multiplication)

2) EF = (xij)mn, where xij = eijfij.

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Thm 7.2 : A: set, |A| = n, R  A  A, let M denote the relation
matrix for R. Then
(a) R is reflexive iff In  M.
(b) R is symmetric iff M = Mtr.
(c) R is transitive iff MM = M2  M.
(d) R is antisymmetric iff MMtr  In.
Proof. (1/2)
Let M = (aij)nn.
(c) () Let M 2  M. If (x, y), (y, z)  R.
 mxy = myz = 1
(mxy means the entry of M in row (x), column (y))
 sxz = 1
(sxz means the entry of M2 in row (x), column (z))
∵ M2  M ∴ mxz = 1
 (x, z)  R and R is transitive.
(c) Spring 2021, Justie Su-Tzu Juan
§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Thm 7.2 : A: set, |A| = n, R  A  A, let M denote the relation
matrix for R. Then
(c) R is transitive iff MM = M2  M.
Proof. (2/2)
(c) () If R is transitive
Let sxz  the entry in row (x) and column (z) of
M2 = 1
∵ sxz = 1 ∴  y  A s.t. mxy = myz = 1
 (x, y)  R  (y, z)  R
 (x, z)  R (∵ R is transitive)
 mxz = 1
∴ M2  M.

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
Def 7.14 : V: finite nonempty. A directed graph (or digraph) G 
 G = (V, E), where V is called the vertex set, E  V  V is called

the edge set.

 v  V is called the vertices or nodes of G

 (a, b)  E is called the (directed) edges or arcs of G

 a is called the origin or source of (a, b)

 b is called the terminus or terminating vertex of (a, b)

 a is adjacent to b; b is adjacent from a

 (a, a) is called a loop at a

isolated vertex
5
Ex 7.25 : V = {1, 2, 3, 4, 5}, 1
2
E = {(1, 1), (1, 2), (1, 4), (3, 2)}
4 3

(c) Spring 2021, Justie Su-Tzu Juan

§ 7.2 Computer Recognition: Zero-One Matrices and
Directed Graphs
a b a b
Def : If (a, b), (b, a)  E, (a  b), then use {a, b} = {b, a} to
represent. a and b are called adjacent vertices.

Ex 7.26 : precedence graph for the computer program

(S1) b := 3
(S2) c := b + 2
(S3) a := 1
(S4) d := a  b + 5
(S5) e := d – 1
(S6) f := 7
(S7) e := c + d
(S8) g := b  f

(c) Spring 2021, Justie Su-Tzu Juan