1

REVISED SYLLABUS
S.Y.B.SC. (I.T.), SEM. -III,
LOGIC AND DISCRETE MATHEMATICS
Unit 1 :
Set Theory : Fundamentals - Sets and subsets, Venn
Diagrams, Operations on sets, Laws of Set Theory, Power
Sets and Products, Partition of sets, The Principle of Inclusion
- Exclusion.
Logic : Propositions and Logical operations, Truth tables,
Equivalence, Implications, Laws of Logic, Normal forms,
Predicates and quantifiers, Mathematical Induction.
Unit 2 :
Relations, diagraphs and lattices : Product sets and
partitions, relations and diagraphs, paths in relations and
diagraphs, properties of relations, equivalence and partially
ordered relations, computer representation of relations and
diagraphs, manipulation of relations, Transitive closure and
Warshalls algorithm, Posets and Hasse Diagrams, Lattice.
Unit 3 :
Functions and Pigeon Hole Principle : Definitions and types
of functions : injective, subjective and bijective, Composition,
identity and inverse, Pigeon hole principle.
Unit 4 :
Graphs and Trees : Graphs, Euler paths and circuits,
Hamiltonian paths and circuits, Planner graphs, coloring
graphs, Isomorphism of Graphs.
Trees : Trees, rooted trees and path length in rooted trees,
Spanning tree and Minimal Spanning tree, Isomorphism of
trees, Weighted trees and Prefix Codes.
Unit 5 :
Algebric Structures : Algebraic structures with one binary
operation - semi groups, monoids and groups, Product and
quotient of algebraic structures, Isomorphism, homomorphism,
automorphism, Cyclic groups, Normal sub group, codes and
group codes, Algebraic structures with two binary operations rings, integral domains and fields. Ring homomorphism and
Isomorphism.

Unit 6 :
Generating Functions and Recurrence relations : Series
and Sequences, Generating Functions, Recurrence relations,
Applications, Solving difference equations, Fibonacci.
Books :
Discrete mathematical structures by B Kolman RC Busby, S
Ross PHI Pvt. Ltd.
Discrete mathematical structures by R M Somasundaram
(PHI) EEE Edition.
Reference :
Discrete structures by Liu, Tata McGraw -Hill.
Digital Logic John M Yarbrough Brooks / cole, Thompson
Learning
Discrete Mathematics and its Applications, Kenneth H. Rosen,
Tata McGraw - Hill.
Discrete Mathematics for computer
Mathematicians, Joe L. Mott, Abraham

scientists

and

Kandel Theodore P. Baker, Prentice - Hall of India Pvt. Ltd.

Discrete Mathematics With Applications, Susanna S. Epp,
Books / Cole Publishing Company.
Discrete Mathematics, Schaums Outlines Series, Seymour
Lipschutz, Marc Lipson, Tata McGraw - Hill.

1
FUNDAMENTALS OF DISCRETE
MATHEMATICS
Unit Structure :
1.0
1.1
1.2

1.3

1.4
1.5
1.6

Objectives
Introduction
Sets and Subsets
1.2.1 Sets
1.2.2 Some basic definitions
Operation on sets
1.3.1 Basic definitions
1.3.2 Algebraic properties of set operations
1.3.3 Principle of Inclusion-Exclusion
Let us sum up
References for further reading
Unit end exercise

1.0 OBJECTIVES

After going through this chapter you will be able to :

Understand basic tools of discrete mathematics like sets.
Understand different type of sets.
Understand different operation on sets.
Principle of Inclusion- Exclusion.
Solve different examples on sets, and principle of InclusionExclusion etc.

1.1 INTRODUCTION
Sets are one of the most fundamental concepts in
mathematics. It was invented at the end of the 19th century. It is
seen as the foundation from which all of the mathematics can be
derived. Concept of division of Integers is fundamental to computer
arithmetic. Mathematical structures are useful in Number theory.
Number theory plays an important role in Computer and Internet
security.

1.2 SETS AND SUBSETS

1.2.1 Sets
A set is any well defined collection of distinct objects.
Objects could be fans in a class room, numbers, books etc.
For example, collection of fans in a class room collection of
all people in a state etc. Now, consider the example, collection of
brave people in a class. Is it a set? The answer is no because brave
is a relative word and it varies from person to person so it is not a
set.
Note : Well-defined means that it is possible to decide whether a
given object belongs to given collection or not.
Objects of a set are called as elements of the set. Sets are
denoted by capital letters such as A, B, C etc and elements are
denoted by small letters x, y, z etc.
If x is an element of set A then we write x A and if x is not
an element of A then we write x A .
There are two ways to represent a set one way by listing all
the elements of a set separated by a comma enclosed in braces.
Another way of specifying the elements of a set is to give a rule for
set membership.
For example, A = e, t, a can be written as
A = x | x is a letter in the word 'eat'

We have following Basic sets of numbers.

(a) = set of all Natural numbers.
= 1, 2, 3,......
(b) = set of all whole numbers
= 0, 1, 2, 3, ......
(c) = set of all Integers
= ...., 2, 1, 0,1, 2, 3,......
(d) = set of all rational numbers.
p
q

= /p, q , q 0

(e) = set of all real numbers.

1.2.2 Some Basic Definitions

(a) Empty Set : A set without any element. It is denoted by or { }
For examples,

B = x | x 1 and x =

C = x | x

and x +1 = 1 =

(b) Equal Sets :- Two sets A and B are said to be equal if they have
same elements and we write A = B.
For examples,
(1) A = x | x is a letter in the word 'ate'
B = y | y is a letter in the word 'eat'

A = B

(2) X = 3, 3 and Y = x | x 2 = 9, x

i.e. X = Y
(c) Subset :- Set A is said to be a subset of B if every element of A
is an element of B and this is denoted by A B or B A . If A
is not a subset of B we write A B .
For example,

(1) A = 1 , B = x | x 2 = 1, x then A B and B A

(2)

Note : (1) Every set A is a subset of itself i.e. A A

(2) If A B but A B then we say A is a proper subset
of B and we write A B . If A is not a proper subset
of B then we write A B .
(3) A for any set A
(4) A = B iff A B and B A
(d) Finite Set :- A set A with n distinct elements, n is called
as a finite set.
For example,
(1) A = x | x , 5 x 2 0

(2) B = y | y is a hair on some ones head

(e) Infinite Set :- A set which is not finite is Infinite.

For example,
(1) A = x | x , 1 x 2
(2)
(3)
Note : is finite.
(f) Cardinality of a set :- The number of elements in a set is called
as cardinality of a set and it is denoted by n(A) or |A|.
For example,
(1) A = {1, 2, 3, 4, 5}, |A| = 5
(2) B = , |B| = 0
(g) Power set :- Let A be a given set. Then set of all possible
subsets of A is called as a power set of A. it is denoted by
P(A).
For example,
(1) A = {1, 2}
P(A) = { , {1}, {2}, {1, 2}}
Note : (1) If |A| = m then |P(A)| = 2m
(h) Universal set :- Any larger set which contains some subsets is a
universal set. It is denoted by U.
For example,
(1) contains , , and .

is a universal set for , ,

and . Similarly, is

universal set for and and so on.

(2) A CPU consist of hard disk, RAM, ROM, Sound Card etc. It
can be treated as a universal set.
(i) Venn diagram :- A pictorial representation of a set is called as
Venn diagram. Elements of a set are denoted by dots enclosed in
a triangle, a square or a circle.

For example,
(1) A = {a, b, d}

(2) B = {5, 6, 7}
A

Fig. 1.1

Fig. 1.2

Check your progress :

1. Identify each of the following as true or false.
(a) A = A (b) A A
(c) A A
(d) A
(e) A (f) If A {1} then P(A) = { , A}
2. If A = { x,y, 3 }, then find (a) P(A), (b) |A| (c) |P(A)|
3. Which of the following are empty sets?
(a) x | x ,1
x 2

(b) x | x , x2 = 1

(c) x | x , x +1 = 1

(d) x | x , x 2 = 3

4. Draw the Venn diagram for .

1.3 OPERATION ON SETS

1.3.1 Basic definitions :
(a) Union of two sets :- Let A and B be two given sets. Union of A
and B is the set consisting of all elements that belong to A or
B and it is denoted by A B . A B = x | xA or x B
For example,
(1) A = x,y,z , B = 2, 5
A B = x,y, z, 2, 5

(2) A = , B =

A B = = 1, 2, 3, ..... ..... , 2, 1, 0,1, 2, ....

= ....., 2, 1, 0,1, 2, .... =

Note : (1) If A B then A B = B

(2) A A = A
(3) A = A
B

Shaded region represents A B .

(4)
Fig. 1.3

(b) Intersection of two sets :- Let A and B be two given sets.

Intersection of A and B is the set consisting of the elements
Present in A and B. (i.e. in both) and it is denoted by A B .
A B = x | x A and x B

For example,
(1) A = {1, 2, 3}, B = {2, 4, 5} then A B = 2
(2) A = {1, 2, 3} and B = { x, y } then A B = such sets whose
intersection is empty is called as disjoint sets.
Note : (1) A A = A
(2) If A B then A B = A
(3) A =
(4) Shaded region represents A B

(5) Disjoint sets

U
A

Fig. 1.4

Fig. 1.5

Definition for union and Intersection can be extended to n

number of sets. ( n )
(c) Complement of a set :- Let U be a given universal set and let A
be any subset of U. Then complement of a set A in U is set of
those elements which are present in U but not in A and it is
denoted by Ac or A or A .
i.e. A c = x | x A & x U

For example,
(1) If A = and U = then A c = 0 but if U = then
A c = ...., 2, 1, 0 i.e. complement of a set depends on U.
U
Ac

Fig. 1.6
Shaded region is Ac .
(d) Complement of A with respect to B is the set of all elements in
B which are not in A and it is denoted by A B or A\ B .
i.e. A B = x | xA and x B
Similarly, we can define B A.
For example,
(1) A = {1, 2, 3} and B = {3, 4, 5}
A B = {1, 2} and B A= {4, 5}
(2) A = {1, 2} and B = {3, 4}
A B = A and B A = B

Fig. 1.7

Fig. 1.8

AB

BA

(e) Symmetric difference of two sets :- Symmetric difference of two

sets A and B is the set of elements present in A or in B but not
both and it is denoted by A B = A B B A .

A+ B
Fig. 1.9

10

Check your progress :

1. If A = U ( U universal set), then (a) A c , (b) A \U , (c) U \ A ,
(d) A U
2. If U = x | x and x17 and A = {1, 3, 5, 6}, B = {3, 4, 7,
5, 8}, then (a) A c , (b) A\B, (c) B\A, (d) A B , (e) Bc , (f) A B ,
(g) A B .
1.3.2 Algebraic Properties of set operations
Like Algebraic properties of Real numbers, sets also satisfy
some Algebraic Properties with respect to the operations union,
intersection etc.
(I) Commutative Properties
(1) A B = B A
(2) A B = B A
(II)Associative Properties
(3) A B C = A B C
(4) A B C = A B C

(III) Distributive Properties

(5) A B C = A B A C

(6) A B C = A B A C

(IV) Idempotent Properties

(7) A A = A
(8) A A = A
(V)

Properties of Complement

(9) A = A
(10)
(11)
(12)
(13)
(14)
(15)

A A = U
A A =
=U
U=
A B = A B
A B = A B

(De Morgans laws)

11

Properties (1) to (13) can be proved easily. We will prove (14) and
(15) here.
(14) A B = A B
Proof : A B = x | x A B and x U

= x | x A and x U and x B and x U

= x | x A and x B
= A B

Similarly, we can prove (15).

Example 1: Prove that i) ( A B) ( A B c ) A and
ii) ( A B) ( A B c ) A .
Solution: L.H.S.= ( A B) ( A B c )
= A ( B Bc )
( Distributive law)
c
= A
( B B complement law)
= A
= R.H.S
Hence ( A B) ( A B c ) A .
Similarly, we can prove ( A B) ( A B c ) A .
Example 2: If U = {x\x is a natural number less than 20} is the
universal
set, A = {1, 3, 4, 5, 9}, B = {3, 5, 7, 9, 12}. Verify that
De Morgans laws.
Solution: De Morgans laws can be state as i) A B = A B ,
ii) A B = A B .
By listing method,
U = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19},
and A = {1, 3, 4, 5, 9},
A = {2, 6, 7, 8, 10, 11, 12, 13,14,15,16,17,18,19},
and B = {3,5,7,9,12},
B = {1, 2, 4, 6, 8, 10, 11, 13, 14, 15, 16, 17, 18,19}
A B = {1, 3, 4, 5, 7, 9, 12}
( A B) = {2, 6, 8, 10, 11, 13, 14, 15, 16, 17, 18, 19}
Also ( A B) = {2, 6, 8, 10, 11, 13, 14, 15, 16, 17, 18, 19}

12

Hence A B = A B .
Now ( A B ) = {3, 5, 9},

( A B) = {1, 2, 4, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19}

Also ( A B) = { 1, 2, 4, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17,

18, 19}
Hence A B = A B .
Example 3: If u={1,2,3,4,5,6,7,8,9,10,11,12} is the universal set.
A={2,3,5,8,10}, and
B={4,5,7,8,9,11}, find.
i). A-B, ii). B-A, iii). ( A B) .
Solution:- i). A-B={2,3,10}
ii). B-A={4,7,9,11}
iii). ( A B) ={1,4,5,6,7,8,9,11,12}.
1.3.3 Principle of Inclusion Exclusion (The addition Principle)
Theorem (1) If A and B be two given finite sets, then we have
A B = A + B A B .
(1) Let A = {a, b, c, d} and B = {e, d, p, q}
A B = {c, d}
i.e. |A| = 4, |B| = 4 and | A B | = 2

By addition principle,

A B = 4 + 4 2 = 6

Example 4: In a class of 50 students, 25 like Maths and 15 like

Physics, 10 like both Maths and Physics. So find (a) How many
like Maths or Physics? (b) How many do not like any or the
subjects?
Solution : Let m be the set of all those students who likes Maths and
p be the set of all those students who likes Physics.
|M| = 25, |P| = 15 and |M P| = 10
(a) No. of students like atleast one subject
= M P = M + P M P (by Addition formula)
= 25 + 15 10
= 30
(b) No. of students do not like any of the two subject
= 50 M P = 50 30 = 20
Above theorem can be extended to three sets,

13

Theorem :- If A, B and C be given finite sets, then

ABC = A + B + C AB BC AC + ABC

Example 5
In a survey of people it was found that 80 people knew
Maths, 60 knew physics, 50 knew chemistry, 30 new Maths and
Physics, 20 knew Physics and Chemistry, 15 knew Maths and
Chemistry and 10 knew all three subjects. How many people knew?
(a) At least one subject
(b) Maths only
(c) Physics only
(d) Maths and Chemistry only
Solution : Let M, P, C represents respectively, the set of students
knowing Maths, Physics and Chemistry.
|M| = 80, |P| = 60, |C| = 50, M P = 30 , M C = 15 ,
P C = 20 , M P C = 10

(a) By addition principle

MPC = M + P + C MP MC PC + MPC

= 80 + 60 + 50 30 15 20 + 10
= 135
Lets draw the Venn diagram of above situation.
M

P
20

5
5

20

10
10
25

Fig. 1.10
(b) Maths only = 80 (20 + 10 + 5)
= 80 35
= 45
(c) Physics only = 60 (20 + 10 + 10) = 20
(d) Maths and Chemistry only = 15 10 = 5
Example 6: Out of 150 residents a building, 105 speak Marathi, 75
speak Gujarati and 45 speak both Languages. Find the number of

14

residents who do not speak either of the languages also find the
number of residents who speak only Marathi.
Solution:- Let A be the set of resident who speak Marathi.
B the set of resident who speak Gujarati
Given U=150
A=105
B=75
AB=45
By principal of Inclusion-Exclusion.
AUB=A+B-AB
= 105+75-45
=135.
i). Number of resident who do not speak either of language.
(AUB)=U-AUB
=150-135
=15
ii). The number of resident who speak only Marathi
=A-AB
=105-45
=60.
Example 7: Out of 240 students in college 130 students are in
N.C.C. 110 are in N.S.S. and 80 are in other activity in this 40 are
N.C.C. and N.S.S both, 35 are N.C.C and other activity and 30 are
N.S.S. and other activity but 20 student are take part in all three.
Find the number of students takes part in
i). Atleast any one.
ii). None of them.
iii). Only N.S.S.
iv). Only other activity.
v). Only N.S.S and N.S.S but not in other activity.
Solution:- Let A be the set of N.S.S students.
B be the set of N.C.C students.
C be the set of other activity student.
Here A=130, B=110, C=80.
AB=40, AC=30, BC=35, ABC=20.
By principle Inclusion-Exclusion
i). atleast one of them i.e.AUBUC
AUBUC=A+B+C-AB-BCAC+ABC
=130+110+80-40-35-30+20
=235.

15

ii). None of them i.e. (AUBUC)

(AUBUC)=U-AUBUC
=240-235
=5.
iii). Only N.S.S. i.e.
=A-AB -AC+ABC
=130-40-30+20
=80.
iv). Only other activity i.e.
=C-AC-BC+ABC
=80-30-35+20
=35.
v). Only N.S.S and N.C.C. but not other activity i.e.
=AUB-ABC
=40-20
=20.
Example 8: Find the number of integers including both from 1 to
500 that are divisible by,
i). 2 or 3 or 5.
ii). 2 or 3 but not 5.
iii). Only by 5.
Solution:- Here U={1,2,3,,500}
A=the set of numbers in U divisible by 2.
B= the set of numbers in U divisible by 3.
C= the set of numbers in U divisible by 5.
500
250
U=250, A=
2
500
166
B=
3
500
C=
100
3
500 500
AUB=

83
2 3 6
500
BC=
33
15
500
50
CA=
25
500 500

16
ABC=
2 3 5 30

16

i) AUBUC A B C A B B C A C A B C
= 250+166+100-83-33-50+16
= 366.
ii). 2 or 3 but not by 5.
=AUB-ABC
=83-16
=67.
iii). Only by 5.
=C-AC-BC+ABC
=100-33-50+16
=33.

1.7 LET US SUM UP

This chapter consist of sets and different operations on sets
with different examples which helps in better understanding of the
concept and able to use in different areas. We saw the principle of
inclusion Exclusion which can be used in different counting
problems we saw some concepts of number theory such as division
in Integers, sequence etc. which is useful in computer security. At
the end we saw definition of a mathematical structure and it is
different properties.

1.8 REFERENCES FOR FURTHER READING

a) Discrete structures by Liu.
b) Discrete mathematics its Application, Keneth H. Rosen TMG.
c) Discrete structures by B. Kolman HC Busby, S Ross PHI Pvt.
Ltd.
d) Discrete mathematics, schaums outlines series, seymour Lip
Schutz, Marc Lipson, TMG.

1.9 UNIT END EXERCISES

1. Let A = x | x and x 2 + 7 = 0 , B = x | x ,
C = x | x , 0 < x < 0.2 , D x | x 6q, q
E = x | x , x + 7 = 7

Check whether following are True or False.

(i) A is finite, (ii) B A , (iii) E = A , (iv) E A D ,
(v) C is infinite, (vi) B = , (vii) A E , (viii) B C = A
2.
Prove A B = A (A B)

17

3.

There are 250 students in a computer Institute of these 180

have taken a course in Pascal, 150 have taken a course in
C++, 120 have taken a course in Java. Further 80 have taken
Pascal and C++, 60 have taken C++ and Java, 40 have taken
Pascal and Java and 35 have taken all 3 courses. So find
(a) How many students have not taken any course?
(b) How many study atleast one of the languages?
(c) How many students study only Java?
(d) How many students study Pascal and C++ but not Java?
(e) How many study only C++ and Java?
4. The students stay in hostel were asked whether they had a
textbook r a digest in their rooms. The results showed that
650 students had a textbook, 150 deed not have a textbook,
175 had a digest and 50 had neither a textbook nor a digest.
Find, i). the number of students in hostel , ii).How many have
a textbook and digest both, iii). How many have only a digest.
5. Prove that (BcU)(AcU )=(AUB)c.
6. Prove that , i).AU(AB)=A, ii). A(AUB)=A.
7. In a survey of 80 people in Gokuldham 50 of them drink Tea,
40 of them drink Coffee and 20 drink both tea and coffee.
Find the number of people who take atleast one of the two
drinks also find the number of students who do not take tea or
Coffee.
8. In a survey of 60 people, It was found that 25 read magazine.
26 read Times of India and 26 read DNA. Also 9 read both
magazine and DNA, 11 read both magazine and times of
India , 8 read times of India and DNA and 8 are not reading
anything.
i). Find the number of people who read all three.
ii). Draw a Vann diagram.
iii). Determine the number of people who read exactly one
magazine.

18

2
LOGIC
Unit Structure :
2.0
2.1

2.2

2.3
2.4
2.5
2.6

Objectives
Propositions and Logical Operations
2.1.1 Logical connectives and compound statements
2.1.2 Negations
2.1.3 Conjunction
2.1.4 Disjunction
Conditional Statements
2.2.1 Quantifiers
2.2.2 Bi-Conditional
Theorem
Mathematical Induction
Let us sum up
Unit End Exercise

2.0 OBJECTIVES
After going through this chapter you will be able to:

Learn about the propositions related to Logical Operations.

Learn about negations and various connectors like conjunction,

disjunction.

Learn about Quantifiers and Conditional and bi-conditional

statements.

Solve problems using the method of Mathematical Induction.

2.1 PROPOSITIONS AND LOGICAL OPERATIONS

The statement in English need not always be true or false
whereas a statement in Logic is a declarative sentence which is
either true or false but not both. Identify the statements among the
following sentences.

19

1. It will rain today.....

2. Mumbai is capital city of Maharashtra.
3. Do you know where is Vijay ?
4. 2 3 5 = 1.
5. x 2 1 = 4.
6. Come in!
In above, sentences (1), (2) and (4) are statements. (3) is not a
statement as it is question,(5) is declarative but depending upon the
value of x it is true or false. Sentence (6) is a command and hence
not a statement.
2.1.1 Logical connectives and compound statements.
Just as in mathematics variables x, y, z, ... can take real values
and can be combined by operations +,,,, in logic, the variables p,
q, r, .. can the replaced by statements. The variable p, q, r, .... are
called as propositional variables. For example we can write p : Sonia
Gandhi is president of India, q : Newton was a Physicist, r : It will
rain today. etc. One can combine propositional variables by logical
connectives to obtain more complex statements - compound
statement. For example suppose Q Mangoes are ripe, R : Oranges
are sour. The statement Q and R means Mangoes are ripe and orange
is sour. The truth value of compound statement depends on truth
values of statements which are combined and on the logical
connectives that are used. In this subsection, we will discuss most
commonly used logical connectives.
2.1.2 Negation :
Suppose P is any statement. Then negation of P, denoted
by p . Thus if P is true then p is false and vice a versa. A table
giving truth values of compound statement in terms of compound
parts is known as truth table.
p
T
F

F
T

Strictly speaking not P is not compound statement as it is unary

operation.
Example 1 Give negation of
1. p: It is hot.
2. q: 2 is a divisor of 5.

20

Solution :
1. ~P : It is not the case that it is hot i.e. it is not hot.
2. ~q: 2 is not divisor of 5. Since q is false, q is true.
2.1.3 Conjunction
The next operation is conjunction. If p and q are two
statement then conjunction of p and q is the compound statement p
and q. The notation is p q . The operation and is a binary
operation on the set of statements. The p q is true whenever both p
and q on true, false otherwise. Thus the truth table is given by
p
T
T
F
F

q
T
F
T
F

pq

T
F
F
F

Example 2 Form the conjunction of p and q.

1. p: I will drive my car q: I will reach the office in time.
2. p: 2 is even q: 11 is odd.
3. p: 2 + 3 + 1 = 6 q: 2 + 3 > 4
4. p: Delhi is capital of India q: Physics is a science subject.
Solution :
1. p q is I will drive my car and I will reach office in time.
2. 2 is even and 11 is odd.
3. 2 + 3 + 1 = 6 and 2 + 3 > 4.
4. Delhi is capital of India and Physics a science subject.
2.1.4 Disjunction
The second logical connective used is disjunction.
Disjunction of statements p and q is dented by p q ,which means p
or q. The statement p q is true where p or q or both are true and
is false only when both p and q are false.
The truth table for p q is as follows.
p
T
T
F
F

q
T
F
T
F

pq

T
T
T
F

21

Example 3 Form the disjunction of

1. p: Bananaras is on the bank of holy river Ganga. q: Dehra Doon is
capital of Uttaranchal.
2. p: Eiffel tower is in London. q: Panama canal connects Atlantic
ocean with Pacific ocean.
3. p: Mukesh and Anil are sons of Industrialist Late Dhirubhai
Ambani.
q: Rajbhavan is official residence of the Governor of
Maharashtra.
4. p:

3 is rational. q: -10 is odd integer

Solution :
The statement p q is given by
1. Banaras is on the bank of holy river Ganga or Dehradoon is
capital of Uttaranchal.
2. Eiffel Tower is in London or Panama canal connects Atlantic
ocean with Pacific ocean.
3. Mukesh and Anil are sons of late Industrialist Dhirubhai Ambani
or Rajbhavan is official residence of the Governor of
Maharashtra.
4. 3 is rational or - 10 is odd.
The statements 1, 2 and 3 are true whereas 4 is false since p
and q are both false.
Note that in logic we can join two totally unrelated sentences
while in English, we do not combine.
In mathematics or in Computer Science, connective or is used
in inclusive sense . That is p q is true if true if p is true or q is true
or both are true. Consider the statements p: 2 is a prime number and
q: 2 is composite. Here the composite statement p q is the
statement 2 is prime number or 2 is composite. Since exactly one
of p and q can be true, is used in exclusive sense.
A compound statement may have many Components each of
which is if self a statement. p q p r involves three
prepositions p, q and r. The prepositions p, q and r each may be
independently true of false. Hence there are in all 23 = 8 possibilities
in the truth table of p q p r . In general, if a statement
involves n propositional variable, then there will be 2n rows in its
truth table.

22

Example 4 Make a truth table if p q q .

p
T
T
F
F

q
T
F
T
F

pq

T
T
T
F

F
T
F
F

F
T
F
F

2.2 CONDITIONAL STATEMENTS

Observe the following sentences that we use in day to day life:1. If it is very hot in summer then there is a chance of early
monsoon.
2. If I see you talking then I will give you a punishment.
3. If I am not in a good mood then I will go for swimming.
4. If I take stress then my blood pressure will increase.
Such sentences are called as conditional statements or implication.
In logic, a compound statement of the type If p then q is called as
conditional statement or implication. p is called as hypothesis or
antecedent and q is called as conclusion or consequent. The
notation for connective if then is denoted by: p q .
Example 5 Write implication for each of the following.
1. p: I have headache q: I will take aspirin.
2. p: I take a walk q: I will reach late.
3. p: 2 divides 10 q: Rajiv will go to movie.
Solution : p q in each of the case is given by
1. If I have a headache, then I will take Aspirin.
2. If I take a walk then I will be late.
3. If 2 divides 10 then Rajiv will go for a movie.
Note than in 1 and 2 given above, we are assuming that p is cause of
q. But in logic, p q means that if p is true then q will also be true.
Hence it is not possible to have p to be true and q is false. Thus
p q is false only when p is true and q is false. In all other
possibilities p q is always true.

23

The truth table for p q is as follows.

p
T
T
F
F

q
T
F
T
F

pq

T
F
T
T

Example 6 Determine truth value of the following :

1. If 3 is even then India will win world cup football.
2. If 2 + 5 = 7 then Sonia Gandhi is the prime minister of India.
Solution:
1. This is true sine 3 is even is false. Note that when p is false, p
q is always true for any q.
2. This is false because 2 + 5 = 7 is true and Sonia Gandhi is PM is
false.
In mathematics we say p implies q or if p then q, q if p, p is
sufficient for q or q is necessary for p. If p q is the implication ;
then the statement q p is called its converse which also is an
implication. The contra-positive statement of the implication p q
is the statement q p .
Example 7 Give contra-positive and converse of the following
statements.
1. If I am hungry then I will eat.
2. If the three sides of a triangle are equal then each angle of the
triangle is of measure 600 .
3. If today is Sunday then I am going for a movie.
Solution
1. Let p be the statement I am hungry and q be the statement I will
eat. The converse is If I will eat then I am hungry while contrapositive is the statement If I will not eat then I am not Hungry.
2. If each angle of a triangle is of measure 600 then the three sides of
the triangle are equal (Converse); If each angle of a triangle is not
of measure 600 then the triangle is not an isosceles (contrapositive).

24

3. If I go for a movie then it is Sunday (converse); If I do not go for a

movie then it is not a Sunday (contra-positive).
Example 8 : Translate each of the following in verbal language.
p: Hardwork.
q: Success.
(i) (~p) (~q), (ii). p ~q, (iii). pq, (iv). ~q~p
Solution : (i). No hardwork and no success.
(ii). Hardwork but not success.
(iii). If there is hardwork then there is a success.
(iv). If no success then no hardwork.
2.2.1 Quantifiers
A set can some times be conveniently denoted as {x | P(x)}
which means that an element x is in the given set if it satisfies the
given property, P(x). For example the set 0, 2, 4, 6,...... can be
described by {x | x is an even integer}. The sentence P(x) is called as
predicate or propositional function. Let A = 0, 2, 4, 6,...... .
Then the sentence P(x) x is an even integer. Since P(2) is true, 2
A while 1 A as P(1) is not true. The universal quantification of a
predicate P(x) is the sentence for all values of x, P(x) is true and
we write this as x P(x). The symbol is called as universal
quantifier. We now discuss few examples.
Example 8 (a) Suppose P(x): 2x + 1 is an odd number is a predicate
that is true for all real numbers x. Hence the sentence x P(x) is
true. (b) Let
Q(x): x 2 9 < 16. In this case x Q(x) is not true since Q(10) is not
true.
A predicate may contain several variables and universal
quantification can be applied to each of them. For example
x y z (xy)z = x(yz) means that the property that the
multiplication is associative is true for all real numbers x,y and z.
The existential quantification of a predicate P(x) is the
statement there exist some value of x for which P(x) is true. We
write this symbolically as x P(x). The symbol is called as
existential quantifier. in English, x can also be described as there
is some x there is at least one value of x.

25

Example 9 Let x and y be any two real numbers.

(a) The statement x y x + y = y is true since real number 0 has
the property that 0 + y = y for all real numbers y.
(b) The statement x y x + y = y is not true since for real number
1 there is no real number y such that 1 + y = y.
Let p: x P(x). Then p will be false if there is at least one value of x
for which P(x) is false. Thus there is at least one value of x for which
P(x) is true. Thus p is false if x P(x) is true. Let q: x Q(x).
Then q is false if there does not exists any value of x for which Q(x)
is true i.e. for all values of x, Q(x) is true. Thus q is false if x
Q(x) is true.
Example 10
(a) Let p: For all integers n, 3n-7 a perfect square. Then p is the
statement. There exists at least one integer n for which 3n - 7 is not a
perfect square.
(b) Let q: there exists a real number x such that 2
is the statement For all real numbers x , 2

x2 1
x2 1

x2 1
x2 1

. Then q

2.2.2 Bi-conditional
Bi-conditional or equivalence of two statements p and q means both
p qandq pand is denoted by p q . The truth table of p q
is given below.
p
T
T
F
F

q
T
F
T
F

r: pq

s:q p

rs

T
F
T
T

T
T
F
T

T
F
F
T

Note that p q is true if either p, q are both true or both false.

Example 11 Compute the truth table of p q q p
p q
T
T
F
F

T
F
T
F

pq

q p

p q q p

T
F
T
T

F
T
F
T

F
F
T
T

T
F
T
T

T
T
T
T

26

Note that last column of the above table indicates that the statement
p q q p is always true. Such a statement is called as
Tautology. A statement that is always false is called as Contradiction
or absurdity. Any other statement is called as contingency.
Check Your Progress :
1) Translate each of the following in verbal language.
a) p: Teacher is present
q: student attend the class.
(i) pq , (ii) pq , (iii).pq, (iv).pq,
b) p: 2 is an even number.
q: 2 is a prime number.

(v). qp

r: 2+2=(2)2.
(i). pq, ii). qp, (iii). pq, (iv). p(q r), (v). ~p q,
(vi). p ~r, (vii). ~p(~p ~r).
2).Write down the following conditional statements in converse
contra positive and inverse.
a). If it is a Sunday then it is a holiday.
b). If the teacher is present then students are standing.
c). If you know mathematics them you know logc.
3) Which of the following statements are Tautology, Contradiction
or Contingency?
1. q p q p
2. p q p q
3. p p
Similar to the mathematical structure [Sets, , ], one can
define a structure on set of proposition with the help of binary
operations , and .
The operations for propositions have following properties
which we list as theorem. The proofs are very simple and hence are
left as an exercise to the reader.

2.3 THEOREM
A. Commutativity
(a) p q q p
(b) p q q p
B. Associativity

27

(a) p q r p q r
(b) p q r p q r
C. Distributivity
(a) p q r p q p r
(b) p q r p q p r
D. Idempotent Property
(a) p p p
(b) p p p
E. Properties of negation
(a) p p
(b) p q p q
(c) p q p q
F. Properties of implication
(a) p q q p
(b) p q p q

(c) p q p q p q
(d) p q p q q r

Standard way of proving all the above properties is to

construct truth table. In some cases one can also use previous results.

2.4 MATHEMATICAL INDUCTION

We now use the ideas developed so far and demonstrate an
important technique of proof- Principle of mathematical induction
which is an indispensable proof technique, extensively used in
mathematics. Suppose P(n) is some statement or property or a
formula to be verified where n is an integer. We need to establish the
formula P(n) is true for all integers n n0 , where n0 is some fixed
integer. This can be achieved as follows. First we establish the
validity of P(n) for n = n0 . This step is called as basis step. Next we
show that P(k ) P (k 1) is a tautology
i.e. assuming the
validity of P(k), we establish the validity of P(k + 1) for any integer
k n9 . This step is called as induction step. Usually some efforts
are required to prove induction step. We now use the induction
principle and prove many formulas, statements.

28

Example 12 Prove that 1 + 2 + 3 + . . . . . . + n

n(n 1)
,n 1.
2

Solution: Let P(n) denote the formula 1 + 2 + 3 + . . . . . +

n

n(n 1)
, . n 1 In this example, n0 1 .
2

Basis step: For n = 1, left hand side of above formula is 1while the
right side is

1(1 1)
which also is 1. Thus the formula is true for n =
2

1 and we have proved the basis step.

Induction step: We now need to establish the validity of the
formula P(k+1) whenever P(k) is true for k 1. Let k 1 be any
integer for which P(k) is true. Thus the statement 1 + 2 + 3 +

k (k 1)
. We now wish to prove 1 + 2 + 3 + . +
2
k 1 k 1 1

. k

k 1

. We can write the right hand side of this as

1 + 2 + 3 + + k + (k + 1). By induction step we know that 1 + 2

+ 3 +.+ k

k (k 1)
.
2

k (k 1)
(k 1). This upon
2
(k 1)(k 2)
straightforward simplification reduces to
. This is
2

Thus 1 + 2 + 3 + .. k (k 1)

precisely what we need to establish. We conclude by the principle of

mathematical induction that P(n) is true for all positive integers.
Example 13. Prove by the method of induction, that for all nN ,
1.2.3 + 2.3.4 + 3.4.5 +..+ n (n + 1) (n + 2) =

nn 1
n 2
n 3
.
4

Solution:- The result P(n) to be proved is, that for all n N

1.2.3 + 2.3.4 + 3.4.5 +..+ n (n + 1) (n + 2) =

nn 1
n 2
n 3
.
4

Step.1: For n = 1, L.H.S = 1.2.3 = 6 ;

1, 2,3, 4
R.H.S =
6
4
There for L.H.S = R.H.S for n = 1
There for P(1) is true
Step.2: Let us assume that for some k N, P(k) is true.
That is, 1.2.3 + 2.3.4 + 3.4.5 + . + k (k + 1)
k k 1k 2 k 3
(I)
k 2
4

29

Then to prove that P (k + 1) is true,

That is, to prove that
1.2.3 + 2.3.4 + 3.4.5 + .. + k (k + 1) (k + 2) + (k + 1) (k + 2)
k k 1k 2k 3k 4
k 3
4
Here
L.H.S = [1.2.3 + 2.3.4 + 3.4.5 + + k (k + 1) (k + 2) ] + (k + 1)
(k + 2) (k + 3)
k k 1k 2 k 3
=
+ (k+1) (k + 2) (k + 3) [by (I).]
4
= (k + 1) (k + 2) (k + 3)

k k 1k 2k 3k 4
4
= R.H.S.
There for P (k + 1) is true.

There for by the principal of mathematical induction, the result P (n)

is true for all n N, that is, 1.2.3 + 2.3.4 + 3.4.5 + + n
(n + 1) (n + 2) =

nn 1
n 2
n 3
, for all n N.
4

Example 14. Prove by the method of induction, that for all n N ,

1
1
1
1
n

.....

3.5 5.7 7.9

2n 12n 3 32n 3
Solution:- the result p (n) to be proved is that for all n N
1
1
1
1
n

.....

3.5 5.7 7.9

2n 12n 3 32n 3
1
3.5
1
1
R.H.S =

32 1 3 3.5
There for L.H.S = R.H.S for n=1 there for P(1) is true.
Step.1. For n=1 L.H.S =

Step.2. Let us assume that for some k N, P (k) is true.

1
1
1
1
k
That is,
.....(1)

.....

3.5 5.7 7.9

2k 12k 3 32k 3
Then to prove that P (k+2) is true,
That is to prove that

30

1
1
1
1
1
k 1

.....

3.5 5.7 7.9

2k 12k 3 2k 32k 5 32k 5
Here
L.H.S

1
1
1
1

.....
2k 12k 3 2k 32k 5
3.5 5.7 7.9
k
1
.. [by (1)]

32k 3 2k 32k 5

1 k
1
1
2 k 2 5k 3

2k 3 3 2k 5 2k 3 32k 5

k 12k 3
k 1

R.H .S
32k 3 2k 5 32k 5

There for P (k + 1) is true.

There for by the principle of mathematical induction, the result P(n)
is true for all n N.
1
1
1
1
n
That is
, for all

.....

3.5 5.7 7.9

2n 12n 3 32n 3
n N.
Example 15.
n n 1
1
2
3
n

.....

3.4.5 4.5.6 5.6.7

n 2 n 2n 3 6 n 3n 4
Solution:- The result P (n) to be proved is that for all n N,
n n 1
1
2
3
n

.....

3.4.5 4.5.6 5.6.7

n 2 n 2n 3 6n 3n 4
1
;
3.4.5
11 1
1
R.H.S =

61.31.4 3.4.5
There for L.H.S = R.H.S for n=1 there for P(1) is true.
Step.1. for n = 1, L.H.S =

Step.2. Let us assume that for some k N , P (k) is true,

31

That is
k k 1
1
2
3
k

.....

3.4.5 4.5.6 5.6.7

k 2 k 3k 4 6k 3k 4
. (I)
Then to prove that P (k + 1) is true,
That is to prove that

k 1k 2
1
2
k
k

3.4.5 4.5.6 k 2
k 3k 4 k 3
k 4k 5 6 k 4k 5
Here L.H.S

1
k
k 1

3.4.5 4.5.6 k 2k 3k 4 k 3k 4k 5

k k 1
k 1
[by (1)]

6 k 3k 4 k 3k 4k 5

k 1
1
k 3k 4 4 k 5

k 1k 2 5k 6

6 k 3k 4 k 5
k 1k 2k 3
k 1k 2

6 k 3k 4 k 5 6 k 4 k 5
R.H .S

There for P (k + 1) is true.

There for by the principle of mathematical induction, the result P(n)
is true for all n N
That is,
n n 1
1
2
3
n

.....

3.4.5 4.5.6 5.6.7

n 2 n 2n 3 6n 3n 4
For all n N.
Example 15. Show that if P(n) given by 1.6 + 2.9 + 3.12 +.+
n(3n + 3) =
+2n+3 is true for n = k then it is true for n = k +
1. Is P(n) true for all n N?

32

Solution:- Let us assume that P(k) is true.

Then 1.6 + 2.9 + 3.12 +.+ k(3k + 3)
=
+ 2k + 3
Now we have to prove P(k + 1) is true.
There for to prove that
1.6 + 2.9 + 3.12 +..+ k(3k + 3) + (k + 1)(3k + 6) = (k + 1)3
+ 3(k + 1)2 + 2(k + 1) + 3
L.H.S = [1.6 + 2.9 + 3.12 +...+ k(3k + 3)] + (k + 1)(3k + 6)
=(
+ 2k + 3) + 3 + 9k + 6
=(
+ 3k + 1) + 3(
) + 2(k + 1) + 3
3
2
= (k + 1) + 3(k + 1) +2(k + 1) + 3 = R.H.S
Hence if P (n) is true for n = k, it is also true for n = k + 1.
When n = 1, n3 + 3n2 + 2n + 3 = 1 + 3 + 2 + 3 = 9
Which is not the same as L.H.S = 1 (6) = 6.
There for P (n) is not true for n = 1.
Hence P (n) is not true for all n N.
Example 16 Prove that a set containing n elements, n 1, has 2n
subsets.
Solution : We will prove the result by induction on the size of the
set, n. Let P(n) denotes the statement : Number of subsets of a set
containing n elements is 2n . In this problem, n0 = 1.
Basis step : Since the only subsets of a singleton set x ={ x1 } are
and { x1 }, the formula is true for n = 1. Induction step: Suppose
P(k), k 1, is true, i.e. any set with k elements has 2k subsets, k 1.
Let X = x1, x2 ,........, xk 1 . Any subset S of X can be classified into
two types : (A) xk 1 S ; (B) xk 1 S . We will count these subsets
separately. If S is any subset of X of type (A) then S' = S { xk 1 } is
a subset of X { xk 1 } and vice-a-versa. Therefore number of
subsets S of the set X of type (A) is same as number of subsets of a
set X { xk 1 }. Since there are k elements in X{ xk 1 }, by
induction there are 2k subsets of X { xk 1 }. Thus there are 2k
subsets of X of type (A). Any subset S of the set X of type (B) is a
subset of X { xk 1 } and vice-a-versa. By induction it follows that
there are exactly 2k subsets of X of type (B). Thus number of
subsets of the set X containing k + 1 elements is 2k 2k 2k 1 . The
result now follows by the method of induction.

33

Example 17 Prove that 1 2n 3n ,n 2 .

Solution : Let P(n) : 1 2n 3n ,n 2 . In this case, n0 2 .
Basis step : Clearly, 1 22 5 32 9 , it follows that P(2) is true.
Induction step: Suppose P(k), k 2 is true. Thus we have
1 2k 3k , k 2 .
Now 1 2k 1 1 2 2k 2(1 2k ) 1 2(3k ) 1, by induction. Clearly,
2(3k ) 1 3k (3k 1) 3k 3k 3k 1 . Thus P(k + 1) is true. By

induction, the result is true for all integers greater than 1.

Example 18 (De Morgans Laws ) Let A1, A2 ,......, An be any n sets.
Prove that
1.

2.

n n
Ai Ai ,n 1.
i 1 i 1
n n
Ai Ai ,n 1.
i 1 i 1

Solution : Let P(n) be the statement that equality holds in(1) for any
n sets.
Basis step : The statement is true for n = 1.
Induction step : Suppose P(k) is true. Thus we have
k
k
Ai Ai ,n 1.
i 1 i 1

The left-hand side of P(k + 1) is

k 1
Ai A1 A2 ....... Ak Ak 1 A Ak 1
i 1
where A A1 A2 ........ Ak . By De Morgans laws for two sets,

A Ak 1 A Ak 1. By induction,
k

A A1 A2 ...... Ak A :
i 1

34

Hence P(k + 1) is true. The result now follow by induction.

Example 19 Prove that 3 | ( n3 n ) for any natural number n.
Solution : Let P(n) denote the statement that 3 | ( n3 n ) for any
natural number n.
Basis step : Clearly, ( 13 1) = 0 is divisible by 3, it follows that P(1)
is true.
Induction step : Suppose 3 | ( k 3 k ), k 1, i.e.( k 3 k ) = 3t, for
some
integer
t.
Since

k 13 k 1 k 3 3k 2 3k 1 k 1 k 3 k

3 k2 k .

By

induction, the first term on the right is divisible by 3 and hence it

follows that the right side is divisible by 3. We conclude the result
by the principle of induction.

2.5 LET US SUM UP

We have learned logical connectives such as negation,
conjunction, disjunction. We also have learned about quantifiers,
conditional and bi-conditional statement. Finally we have learned a
method of proof the principle of mathematical induction.

2.6 UNIT END EXERCISE

Prove the exercise number 1 to 8 by the principle of induction.
n

1.

i2

i 1

n(n 1)(2n 1)
6

2.

n(n 1)
i3 2
i 1
n

3.

a.r i a.

i 1
n

4.

r n 1
,r 1
r 1
n

2i n(n 1), (2i 1) n2

i 1
n

5.

i 1

(4i 3) (2n 1)(n 1)

i 1

35

6. Prove that sum of first n terms of an arithmetic progression a, a +

d, a +
2d, ..+ [a(n 1)d] is given by
7.

n 2a (n 1)d
2

1
1
1
1
n

........

.
1.2 2.3 3.4
n(n 1) n 1

8.

1 2
3
n
n2

.......
2
.
2
2
n
2 2
2
2
2n

9. Find a formula for 12 32 52 ........ (2n 1) 2 and prove it by

induction.
n

n
1
10. Prove that 1 1 .
2
2

11. Prove that n 2n ,n 1.

12. Let A1, A2 ,......., An and B be any sets. Prove by induction the
following distributive properties
(a)

n
n

Ai B ,n 1.

i
i 1
i 1

(b)

n
n
Ai B Ai B ,n 1.
i 1
i 1

13. If A and B are two square matrices of order n such that AB =

BA. Prove that AB An B n ,n 1. .
n

14.

(a) Prove that product of any two consecutive integers is

divisible by 2.
(b) Prove that product of any three consecutive integers is
divisible by 6.

36

3
RELATIONS AND ITS PROPERTIES
Unit Structure
3.0
Objectives
3.1
Introduction
3.2
Product sets and partitions
3.2.1 Product sets
3.2.2 Partitions
3.3
Relations and diagraphs
3.3.1 Definition and examples of relation
3.3.2 Sets related to a relation
3.3.3 The matrix of a relation
3.3.4 The diagraph of a relation
3.4
Paths in relations and diagraphs
3.4.1 Paths in a relation R can be used to define new
relations
3.4.2 Matrix version
3.5
Properties of relations
3.5.1 Reflexive and Irreflexive relations
3.5.2 Symmetric, Asymmetric and Antisymmetric relations
3.5.3 Transitive relations
3.6
Let us sum up
3.7
References for further reading
3.8
Unit end exercise

3.0 OBJECTIVES:
After going through this chapter you will be able to :
Understand the concept and definition of product and
partition of a set.
Understand the different representation of a relation (set
theoretical, pictorial and matrix representation).
Understand the definition of a path in a relation and able to
find paths of different length.
Understand the different properties of binary relation.

37

3.1 INTRODUCTION:
In day today life we deal with relationships such as an
employee and employee number, element and set, a person and his
telephone number etc. In mathematics its looked in more abstract
sense such as division of integers, order property of Real numbers
and so on. In computer science, a computer programme and variable,
computer language and valid statement and so on Relations are
useful in computer databases, networking etc.

3.2 PRODUCT SETS AND PARTITIONS:

3.2.1 Product sets
Definition:
Let A and B be two non empty sets. The product set or
Cartesian product of A and B, (denoted by A B) is the set of all
ordered pairs (a, b) where a A and b B .
Thus, A B (a, b) / a Aandb B
[ Note: an order pair (a, b) is the ordered collection that has a and
b in prescribed order, a in first position and b in second
position.]
Examples:
(1) Let A = 1, 2, 3 and B = x, y
then A B (1,x),(1,y),(2,x),(2,y),(3,x),(3,y)
Similarly, B A = (x,1), (y,1), (x, 2), (y, 2), (x, 3), (y, 3)
(2) Let A be the set of all 2 divisions in Xth class in some school and
B be the set of all 3 courses available.

i.e. A = X, Y , B = C ++ , Java, VB
then

A B = (X, C++ ), (X, Java), (X, VB) (Y, C++ ), (Y, Java), (Y, VB)

so there are total 6 categories possible.

Remark:
(1) A B and B A may or maynt be equal.
(2) If A and B are finite sets then A B A . B B A

38

(3) An ordered pair (a1, b1 ) and (a 2 , b 2 ) are equal iff a1 a 2 and

b1 b 2 .

(4) The idea of Cartesian product of two sets can be extended to

n number of sets A1,A 2 ,...,A n , (its denoted by
A1A 2 ...A n ) and its defined as, A1A 2 ...A n

(a1,a 2 ,...,a n ) / a1 A1,a 2 A 2 ...&a n A n

[ (a1,a 2 ,...,a n ) is an ordered n -tuple.]

3.2.2 Partitions
Definition:
A partition of a nonempty set A is a collection
p A1,A 2 ,...,A n of nonempty subsets of A such that
(1) A1 A 2 A3 ... A n A
(2) Ai A j 0 (1 i j n)
A1,A 2 ,A3 ,...,A n are called as blocks or cells of the partition.

Example:
(1) A 1,2,3,4,5
(a) Let A1 1 ,

A 2 and A3

Then we have, A1 A 2 A3 A
&A1 A 2 O ,A1 A3 O ,A 2 A3 O

P A1,A 2 ,A3 is a partition of A

(b) Let A1 1,2 ,

A 2 , A3 and A 4

then P A1,A 2 ,A3 ,A 4 is a partition of A.

(c) Let A1 1,2 ,

A 2 , A3

then P A1,A 2 ,A3 is not a partition of A

A1 A 2 O

(d) Let A set of all integers

E = set of all even integers and

39

O = set of all odd integers

We have, E O O and E O A
P E,O is a partition of A.

Check your progress

1.

List all partition of A a,b,d.

2.

Let A a,b,c,d,e,f ,g,h andA1 a , A 2 b,c , A3 d, e, f ,

A 4 g,h andA5 f ,g ,A 6 a,b,c

Which of the following are partition of A.

(a) A1,A 2 ,A3 (b) A1,A 2 ,A3 ,A 4 (c) A3 ,A 4 ,A5 ,A 6
(d)
3.

A3 ,A6 ,A 4

A B B A if (a) A is finite

(b) A=B

(c) B is finite
4.

If A x,y,z ,B 1,2,3andC a,b write down the set

A B C.

3.3 RELATIONS AND DIAGRAPHS:

3.3.1 Definition and examples of Relation
Definition:
Let A and B be two non empty sets A relation R from A to B
is a subset of A B.
If (x, y) R then we write xRy and If (x, y) R then we
write xRy .
Examples:
(1) Let A 1,2,3&B x,y,z then R (1,x), (1,y), (2,z) is
relation from A to B.
Note: If A=B then instead of saying a relation from A to B we will
say a relation on A.
(2) Let A B 1,2,3,4 Let R be a relation on A defined as xRy
iff x > y.
R (2,1),
(3,1),
(3,2),(4,1),
(4,2),(4,3)

40

(3) Let A B The set of Real numbers.

Let R be a relation on A such that
xRy iff x 2 y 2 25

Fig. 3.1

R (x,y) / x 2 y 2 25

= The set of all points on the circle centre at origin with

radius 5
We can see, (3,4) R

( 32 42 25)

but (3, 3) R ( 32 32 18 25)

(4) Let A , Let R be a relation on A defined as xRy iff x
divides y. R (1,2),(2,4),(5,10),
(2,6),....
We have 1 R 2 but 2 R 1 .
3.3.2 Sets related to a relation
Let R be relation from A to B.
Two important sets related to R are the Domain of R
[denoted by Dom (R)] and The Range of R [denote by Ran (R)].
We have, Dom (R)= x /(x,y) R A i.e. Dom (R) is a
subset of A containing first element of the pair (x, y) which
belongs to R. Similarly, Ran (R)= y /(x,y) R B .
For Example:
(1) Let A 1,2,3 and B x, y R (1,x),(3,y)
Dom (R)= 1,3 A and Ran (R)= x,y B
(2) Let A and B be only two sets and R = A B.
Then Dom (R) = A and Ran (R) =B

41

Check your progress

1.

Write down the elements of R form

A 0,1,
2,3

to

B 1,2,3 , defined as (a, b) R iff

2.

(a) a = b
(b) a + b is an even number
(c) a + b is a multiple of 3 (d) a b
Find the domain and Range of the relations defined in Q.1.

3.3.3 The matrix of a Relation

A relation between two finite sets can be represented by a
Boolean matrix (a matrix which is having entries as 0 or 1)
Let R be a relation from A a1,a 2 ,..., a m to B b1,b 2 ,...,b n .
(Here the elements of A and B are listed in a particular order). Then
relation R can be represented by the m x n matrix M R [mij]m n ,
which is defined as,
1if ( ai ,b j )R
mij
0 if ( ai ,b j )R

The matrix MR is called as the matrix of a Relation R

Examples:
(1) Let A = 1, 2, 3 and B x,y and
R (1,x),(2,x),(3,y),(1,y),(3,x)

1 1
M R 10
1 1

3 2

(2) Let A B 1,2,3,4

Let R be a relation on A defined as xRy iff x y .
R (1,1),
(1,2),(1,3),(1,4),(2,2),(2,3),(2,4),(3,3),(3,4),(4,4)

1
0
MR =
0

1 1 1
1 1 1
0 1 1

0 0 1 44

Note: Converse process is also possible i.e. given a matrix with

enteries 0 or 1 we can write R related to that matrix.

42

3.3.4 The Diagraph of a Relation

Just we saw that a relation on finite set A can be represented
by a binary matrix. Similarly there is another way of representing a
relation using a pictorial representation. Pictorial representation of
R is as follows, Draw a small circle for each element of A and
label the circle with the corresponding elements of A, (these circles
are called as vertices) draw an arrow from vertex a i to a j iff
a i R a j . (these arrows are called as edges)

The resulting pictorial representation is called as Directed

graph or diagraph of R.
For example:
(1) Let A B 1,2,3,4 and R (1,1),
(1,2),(2,3),(3,4),(2,4),(3,1)

1

Fig. 3.2
Note:
(1) An edge of the form (a, a) is represented using an arc from the
vertex a back to it self. Such an edge is called a loop.
(2) Conversely diagraph can be used to find underlying relation
represented by it.
(3) There are two important definitions arising from the diagraph.
(i) In-degree of
= no. of arrows
a vertex
coming towards that vertex
and
(ii) Out-degree of = no. of arrows
a vertex
going away from that vertex

43

For example (1) for below diagraph

1
4
2
5
3

Fig. 3.3
1

In-degree

Out-degree

Check your progress

1. Write down the matrix M R and draw the diagraph for following
relations.
(a) A B 1,2,3,4 , R is such that xRy iff x | y.
(b) A B 1,2,3,4 , R is such that xRy iff x y 5
(c) A B 1,2,4,6 , R is such that xRy iff x y is a
multiple of 2
2.

Write down In-degree and out-degree for each of the vertices in

example (1).

3.

Let A , Give a description of the relation R specified by

the shaded region shown below
(0, 2)

(-2, 0)

(2, 0)
(0, -2)

Fig.3.4

44

3.4 PATHS IN RELATIONS and DIAGRAPHS:

Let A be a given set and let R be a relation on A. A path of
length m in R from a to b is a finite sequence.
: a x 0 ,x1,x 2 ,x 3 ,...,x m 1,x m b, starting from a and ending
to b such that aRx1,x1Rx 2 ,x 2 Rx 3 ,...,x m 1Rb.
Note: Length of a path is nothing but the number of arrows involved
in a path.
For example:
(1) Consider the following diagraph,

3
Fig. 3.5
1 :1,2,4 is a path of length 2
2 :1,2,4,1
is a path of length 3
3 : 4,4,3 is a path of length 2
4 : 4,4 is a path of length 1

Note: a path like 2 & 4 are called as cycles, a cycle is a path

which is having same starting and ending vertex.
3.4.1 Paths in a relation R can be used to define new relations.
From above example, It can be seen that paths of length 1 can
be identified with the elements of R and vice versa. So R can be
replaced by R1 where 1 stands for set of all order pairs (x, y) for
which there exist a path of length 1 from x to y. on similar lines now
we can define R 2 ,R 3 ,....,R n (n )

45

R 2 (x,y) / if a path of length ' 2 ' from ' x ' to ' y '.

R n (x,y) / if a path of length 'n ' from ' x ' to ' y '.

Now we may define R as,

R (x,y) / if some path from ' x ' to ' y '.

i.e. R R i
i 1

For example:
(1) Consider the following diagraph.

3
Fig. 3.6
So, R1 (2,2),(2,1),
(2,4),(3,2),(3,4),(1,4),(1,3)
Now for R 2 , we have to find all Paths of length 2
1 : 2,2,2

2 : 2,2,1

3 : 2,2,4

4 : 3,2,2

5 : 3,2,4

6 : 3,2,1

7 :1,3,2

8 :1,3,4

9 : 2,1,
3

i.e. R 2 (2,2, ),(2,1),

(2,4),(3,2),(3,4),(3,1),
(2,3),(1,2),(1,4)
similarly we can find R 3 ,R 4 and so on.

46

(2) Consider the diagraph,

b

Fig. 3.7
Then, R1 (a,b),(b,
a),(b,c),(c,d)
R 2 (a,a),(a,c),(b,b),(b,d)
R 3 (a,b),(a,d),(b,a),(b,c)

R (a,a),(a,b),(a,c),(a,d),(b,a),(b,b),(b,c),(b,d),(c,d)

Note: Since n is finite the process of finding R1,R 2 ,R 3 ,... will

stop after some finite n.
In fact we can prove that, R R1 R 2 R 3 ..... R n ,
where n is number of elements in the given set A.
3.4.2 Matrix version
If R is large, it would be tedious to compute R , or even
R 2 ,R 3 etc. from the set representation of R so we have following

matrix version of above concepts.

First we will see some different operations defined on Boolean
matrices.
Let A a ij and B bij be two m x n Boolean matrices.
(1) we define A B C [Cij ], the join of A and B by,
bij 1
1if a ij 1or
Cij
0if a ij &bijbothare0

47

(2) We define A B E [eij ], the meet of A and B

1if a ij 1& bij 1
eij
0if a ij 0orbij 0

(3) Let A a ij
and B bij
be two Boolean matrices.
m p
p n
Then Boolean product of A and B, (denoted by A B) is the
m n Boolean matrix C [Cij ] defined by

Cij

if aik = 1 and bkj =1 for some k, 1 k p

otherwise

1 0 0
1 0

for eg: Let A 0 1 0

&B 11
1 1 0
0 1
33
32
1 0
A B 11
1 1

3 2

Note:
ai1
a
21
M
a
i1
M
am1

C11 C12 L C1n

b11 b12 K bij L b1n

....
M

M
M
M

M
b
M
M
M
2j

Cij
M
e
= M

ai2 .... aip

M
M
M
M

M
M

b
b
L
b
L
b

M
p1
p2
pj
pn

pxn C
m1 Cm2 L Cmn mxn
am2 .... amp
mxp
ai2
a22

....

aip
a2p

ai1

bij

ai2

b2j

aip

M
bpj

If any corresponding
pair of entries
are both equal
to 1, then C ij = 1 ;
otherwise C = 0
ij

48

We can prove that M

R2

M R M R i.e. matrix related with

R 2 is nothing but the Boolean product of M R with M R similarly.

M n M R M R .... M R (n times)
R

For example
1

Fig. 3.8
as we saw, R 2 (2,1),
(2,2),(2,4),(3,2),(3,4),(3,1),
(1,2),(2,3),(1,4)
0
1
M 2
R
1

1 0 1
1 1 1
1 0 1

0 0 0 44

0
1
and M R
0

0
1
Now lets compute, M R M R
0

0
0
1

i.e. M

R2

0 1 1
1 0 1
1 0 1

0 0 0 44

0 1 1 0
1 0 1 1

1 0 1 0

0 0 0 0

0 1 1
1 0 1
1 0 1

0 0 0

1 0 1
1 1 1
whichisnothingbutM 2
R
1 0 1

0 0 0

M R M R isverified.

49

Notation: M R M R is denoted by M R
2

n
Similarly, M R M R ... M R M R

'n 'times

i.e. M

M R ,n 2
n

Rn

Now, we know that, R R i

i 1

i.e. R R1 R 2 R 3 R 4 ...
we can check that If R and S are two relations then
M R S M R vMS .

if we extend this idea we have,

M M 1 vM 2 vM 3 vM 4 v...
R
R
R
R
R

i.e. M

thus,

M 1 v M R v M R v M R v...
R
2

we

another
way
of
calculating
M R 2 ,M 3 ,.....,M n ,...,M R and which in turn gives the sets
R

got

R ,R ,..., R ,....,R .
2

Check your progress

1. Consider the following diagraph and answer following
1

Fig. 3.9

50

2.

(a) List all paths of length 2

(b) List of all paths of length 2 starting from 3
(c) List all cycles
(d) All cycles starting at 1
For example (1) find M 2 and M 3 ?

3.

Prove that if R and S are two relations then M R S M R vMS

3.5 PROPERTIES OF RELATIONS:

3.5.1 Reflexive and Irreflexive Relations
A relation on set A is reflexive if (x, x) R x A
[orxRx x A]

A relation on set A is irreflexive if

(x, x) R x A

[orxRx x A]

For e.g. (1) Let A 1,2,3,4 with Relations R,S,T on A.

If R (1,1),
(1,2),(2,2),(3,3),(4,4),(4,2) then R is reflexive.
If S (1,1),
(2,1),
(3,3),(4,3),(4,4) then S is not reflexive.
( 2 R 2) and also S is not irreflexive. (3,3) Ror(2, 2) R .

if T (1,2),(2,3),(3,1)
then T is irreflexive.
Note:

(1) (x,x) / x A is called as an equality relation on A.

(2) R is reflexive iff R.
(3) R is irreflexive iff R O
(4) If R O , an empty relation then R is not reflexive since
(x,x) R x A. However, R is irreflexive.
(5) Let R be a reflexive relation on set A then matrix of relation
M R must have diagonal elements as 1.
(6) If R is irreflexive then M R must have diagonal elements as
zeros.

51

3.5.2 Symmetric, Asymmetric and Antisymmetric relations

(1)

relation

on

set

is

symmetric

if

whenever

asymmetric

if

whenever

antisymmetric

if

whenever

(a,b) R,then(b,a) R .

(2) A

relation

on

set

is

(a,b) R,then(b,a) R .

(3) A

relation

on

set

is

(a,b) R,&(b,a) R then a = b.

For Examples:
(1) Let A , R be a relation on A such that xRy iff x divides
y.
(a) If xRy (i.e. x divides y)
then yRx or y R x
(' y ' may or maynt divide x)

For eg: 2 R8 (as 2|8) but 8 R 2( 8 2)

Risn ' tasymmetric

(b) If a = b = 2 then aRb as well as bRa

R is not asymmetric
(c) If a and b are such that a/b and b/a
A / b and b / a gives a = b
R is antisymmetric
(2) Let A = set of all lines in a xy-plane.
(a) If and m are in A such that R m then
m R (Rm m m mR )
R is symmetric

(b) R is not asymmetric as Rm mR

(c) R is not antisymmetric as we can have 2 distinct lines | | to each
other butnt equal.
R
m

Fig. 3.10

52

(3) Let A , R is a relation such that x R y iff x < y then R is not

symmetric but R is asymmetric. ( xRy x y
y x
y R x)

Notes:
1) The matrix M R of a symmetric relation satisfies the property that
MR M

t
R

i.e. if mij 1 then m ji 1 and if mij 0 then m ji 0

2) The matrix M R of an asymmetric relation satisfies the property if

mij 1 then m ji 0 and mii 0 i (i.e. diagonal elements are
zero)
3) Relation R is antisymmetric means x R y and y R x x y
contrapositive of this statement is, if x y xRyoryRx
i.e. M R of antisymmetric relation satisfied the property that if
i j, then Mij 0orM ji 0
Similarly for diagraphs we have,
4) The diagraph of symmetric relation has the property that if there
an edge from i to j, then there is an edge from j to i.
5) If R is an asymmetric relation, then if there an edge from i to j so
there cant be any edge from j to i and there cant be any cycle of
length 1.
6) If R is an antisymmetric relation, then for different i and j there
can not be an edge from vertex i to vertex j and an edge from
vertex j to vertex i. (we cant say anything if i = j )
3.5.3 Transitive Relations
A relation R on set A is said to be transitive if (x, y) R
and(y, z) R then (x, z) R.
[i.e.if xRyandyRz xRz]

For Example:
(1) Let A 1,2,3,4andletR (1,1),
(1,2),(2,2),(2,4),(1,4) and
S (3,2),(2,1),
(1,4),(4,2),(1,2) then

we can check R is

transitive but S is not. [ (3,2)&(2,1)

Sbut(3,1)
S]

53

(2) Let A &Let'R 'bearelation . Then, if xRy and yRz i.e.

x y&y z

We have x y z xRz
R is transitive.

Notes:
(1) A

matrix

M R of

relation

has

the

property,

if

mij 1&m jk 1

Then mik 1
(2) from above point (1) we can see that if M R = m R then R is
2

transitive but converse is not true.

(3) If there is a path of length 2 from a to c then there has to be
a path of length 1 from a to c inorder to have R transitive.
i.e. if (a, c) R 2 then(a, c) R

i.e. R 2 R
R is transitive iff R 2 R

Fig. 3.11

(4) more generally we have,

R is transitive iff R n R

n 1

Check your progress

1. Determine whether the following relation is reflexive,
irreflexive, symmetric, asymmetric, antisymmetric or transitive.
Let A 1,2,3,4
(a) R1 (1,1),
(2,2)
(b) R 2 (1,1),
(2,3),(3,2),(2,2)
(c) R 3 (2,3),(3,2),(2,2),(3,4),(4,3),(3,3),(4,4)
(d) R 4 (2,3),(3,2),(1,4),(4,2),(1,2)

54

2. Determine whether the relation R on the set A 1,2,3 whose

matrix M R and Diagraph are given is reflexive, irreflexive,
symmetric, asymmetric, antisymmertic or transitive.
0 1 0
(a) 1 0 1
0 1 0
(b)

Fig. 3.12
(c)

(d)

1 0 0
0 1 1

1 0 1

Fig. 3.13

3.6 LET US SUM UP

We started the definition of product sets which is useful in
defining relation from one set to other. Then we saw different ways
of representing a relation. Which is useful in understanding the
concepts in more better way. Then we saw very important definition
of a path in a relation and then concepts of paths of different length
and then finally matrix version of it. At the end we saw different
properties of a relation which is useful in coming chapters.

55

3.7 REFERENCES FOR FURTHER READING:

(1) Discrete structures by B. Kolman HC Busloy, S Ross PHI Pvt.
Ltd.
(2) Discrete mathematics and its application, Keneth H. Rosen
TMG.
(3) Discrete structures by Liu.

3.8 UNIT END EXERCISES:

1.

Find R for the relation R whose diagraph is

1

Fig. 3.14
Q.2 Calculate M

R4

for a relation R (1,1),

(3,2),(1,4),(2,4) on set

A 1,2,3,4

Q.3 Determine whether following relations are reflexive,

irreflexive, symmetric, asymmetric, antisymmetric or transitive.
(a) A ,
x + y is an even number.
xRy iff
(b) A ,

xRy iff

x 2 y2 9

(c) A ,

xRy iff

xy

(d) A ,

xRy iff

(x y) 3

Q.4 Define a relation on A 1,2,3,4 that is

(a) Reflexive butnt symmetric
(b) Transitive butnt reflexive
(c) Antisymmetric and reflexive
(d) irreflexive and transitive
Q.5 Prove that if R is symmetric then R 2 is also symmetric.

56

4
EQUIVALENCE RELATION AND
CLOSURES
Unit Structure
4.0 Objectives
4.1 Introduction
4.2 Equivalence relations
4.2.1 Definition
4.2.2 Equivalence relation and partition
4.3 Operations on relations
4.4 Closures
4.4.1 Reflexive and Symmetric closures
4.4.2 Transitive Closure
4.5 Composition
4.6 Computer representation of relations and diagraphs
4.7 Let us sum up
4.8 References for further reading
4.9 Unit end exercises

4.0 OBJECTIVES :
After going through this chapter, students will be able to
Understand the definition of an equivalence relation and able to
identify an equivalence relation
Find the Partition produce by an equivalence relation and vice
versa.
Understand different operations that can be performed on
different relations which is useful in finding the closures of a
relation.
Use Warshalls algorithm to find transitive closure.

57

4.1 INTRODUCTION :
We have already seen the concept of reflexive, symmetric and
transitive etc. If a relation is not transitive then it doesnt contain all
the pairs that can be linked so we want to make it transitive by
adding the remaining pairs with the property that resulting set is
smallest set containing given relation. Such a set is called as
transitive closure.
Also we will see important concept of
composition of relations and computer representation of relation and
diagraph.

4.2 EQUIVALENCE RELATIONS

4.2.1 Definition :
Let R be a relation on A, R is said to be an equivalence relation
on A iff R is reflexive, symmetric and transitive.
Examples :
(1) Let A = {1, 2, 3} and R = {(2, 2), (1, 2), (2, 1), (1, 1), (3, 3)} then
its easy to check R is an equivalence relation.
(2) Let A = and R be a relation on A such that x R y iff x + y is
an even number.
To check whether R is an equivalence relation or not.
(a) Let x A ,

x x 2 x even
x R x x A
R is reflexive.

(b) Let x, y A such that x R y i.e. x y even i.e. x y 2 p, p

Consider, y x x y 2 peven

y x is also an even number.

yRx
R is symmetric.
(c) Let x, y,z A such that x R y x y 2 p, p -------------- (1)
and y R z y z 2q, q --------------- (2)

58

(1) + (2) x 2 y z 2 p 2q

x z 2 p 2q 2 y
x z 2 p q y
i.e. x z even number

xR z
R is transitive.

and hence R is an equivalence relation.

(3) Let A = and R be a relation on A such that x R y iff x
divides y.
We have 3/6 but 6 X 3.
i.e. 3 R 6 but 6 R 3.

R; isnt symmetric,
R isnt an equivalence relation.
4.2.2 Equivalence Relation and Partition
Using an equivalence relation we can produce a partition and
vice versa. Let R be a given equivalence relation on set A.
Let a A
We define a set which is equal to x / x R a, x A is called as
an equivalence class of a. i.e. [a] = x / x R a, x A .
For example,
(1) Let A = {1, 2, 3, 4} and R = {(1, 1), (1, 2), (2, 1), (2, 3), (3, 2),
(2, 2), (4, 4), (1, 3), (3, 1), (3, 3)}
Then, [1] = x / x R 1 = 1, 2, 3
[2] = {1, 2, 3} and [4] = {4}
Theorem :Let R be an equivalence relation on A.
following properties

Then we have

(i) A [ a]
a A

(ii) Any two equivalence classes are equal or disjoint.

59

Proof :(i) T.P.T. A [ a]

a A

[a] = x / x R a, x A A

U [a] A

(1)

aA

Now, let x A t.p.t. x U [a]

aA

x A x R x ( R is reflexive)
x x U [a]
aA

x U [a]

(2)

aA

By (1) and (2), A = U [a]

aA

(ii) Let a, b A, t.p.t. [a] [b] = 0 or [a] = [b]

If [a] and [b] are disjoint then done. So lets consider
[a] [b] 0
Let w [a ] [b]
w [a ] & w [ b]

w R a and w R b

a R w and w R b ( R is symmetric)
a [ b] (by definition)
[a ] [ b] ( a [ b]

a R b and hence every element of

[a] is related to b)

Similarly, we can prove, b a

[a] = [b]
Two equivalence classes are either disjoint or equal.

So,
from above theorem we can say p = [a ] / a A form a
Partition of set A.

Note : Above p is denoted as A/R.

Example
(1) Let A = {1, 2, 3, 4} and R = {(1,1), (1,3), (3,1), (2,2), (4,4)}
[1] = {1, 3}, [2] = {2}, [4] = {4}

p = {[1], [2], [4]} is a Partition of A.

60

We can reverse the above Process, i.e. given a Partition we

can produce an equivalence relation on a given set. Let A be a
given set and let p be a partition of A. define a relation on set A as,
x R y iff x and y belongs to same block or cell of the given
partition.
T.P.T. R is an equivalence relation.
(a) If x A , then its very obvious x belongs to same block.

x R x, x A
R is reflexive.

(b) If x, y A such that, x R y x & y belongs to same block

y & x belongs to same block
yR x

R is symmetric.
(c) If x, y, w A such that,

x R y

& y R w ( x and y belongs to same block)

and (y and w belongs to same lock)

x and w belongs to same block
xRw

R is transitive

R is an equivalence relation.
For example,
(1) Let A = {1, 2, 3, 4} and partition p = {{1}, {2, 3}, {4}}

R = {(1, 1), (2, 2), (2, 3), (3, 2), (3, 3), (4, 4)}

Check your progress

1.

Let A = 1, 2, 3, 4, 5, 6, 7, 8 and p = 1 , 2, 3 , 4 , 5, 6, 7 , 8 be a
partition of A. Find corresponding equivalence relation.

2.

Let

A = 1, 2, 3, 4, 5

and

R = 1,1 , 2, 2 , 1, 2 , 2,1 ,

3,3 , 4, 4 , 3, 4 , 4, 3 , 5,5 be an equivalence relation on

A. Find the corresponding Partition.

61

4.3 OPERATIONS ON RELATIONS :

Let R and s be two relations from set A to set B. We
have following operations defined on Relations.
(1) Complementary relation of R (denoted as R )
R=

x, y / x, y R

(2) Inverse of a relation R (denoted as R 1 )

R 1 = x, y / y, x R

For example,
(1) Let A = 1, 2, 3 and R = 2, 2 , 1, 3 , 2, 3 , 3, 3
R = 1,1 , 1, 2 , 2,1 , 3,1 , 3, 2 and
R 1 = 2, 2 , 3,1 , 3, 2 , 3, 3

(3) R S = x, y / x, y R or x, y S
(4) R S = x, y / x, y R & x, y S
For example,
(1) Let A = 1, 2, 3, 4 and R = 1, 2 , 1,1 , 2, 4 , 3, 2 and
S = 2, 2 , 1,1 , 3, 2 , 3, 4

R S = 1,1 , 1, 2 , 2, 2 , 3, 2 , 3, 4 , 2, 4 and
R S = 3, 2 , 1,1

Note :(1) M R S = M R MS
M R S = M R MS

M R 1 = M R

MR = MR

[ M R complement is a matrix obtained by replacing every 1 by

0 and every 0 by 1.]

62

Theorem : Suppose R and S are relations from A to B

(a) If R S , then R 1 S1
(b) If R S , then S R
(c) R S

= R 1 S1 and R S

= R 1 S1

(d) R S = R S and R S = R S
(e) R is reflexive iff R is irreflexive.
(f) R is symmetric iff R = R 1
(g) R is antisymmetric iff R R 1
(h) R is asymmetric iff R R 1 0
(i) R S R 2 S2
2

Proof :
(a) Let x, y R 1
y, x R S
x, y S1

R 1S1

(b) Let x, y S
x, y S
x, y R
x, y R

R S

S R

(c) R S 1 = x, y / y, x R S
=

x, y / y, x R

x, y / x, y R 1 and x, y S1

= R 1S1
Similarly, other one.

and y, x S

63

(d) R S = x, y / x, y R S
=

x, y / x, y R

or x, y S

x, y / x, y R or x, y S
= x, y / x, y R S
=

= R S
(e), (f), (g) and (h) are easy to check

x, y R S2 means w such that x R S w and

w R S y

(i) Let

x R S w x, w R S and
w R S y w, y R S

i.e. x, w and w, y R and x, w and w, y S

x, y R 2 and

x, y S2

x, y R 2 S2

R S2 R 2 S2
Check your progress
1.

Verify M R S = M R MS , M R S = M R MS , M R 1 = M R ,
T

M R = M R for relations.

2.

R = {(1, 1), (1, 2), (2, 4), (3, 2)} and S = {(2, 2), (2, 4), (1,1),
(4, 3)} on set A = {1, 2, 3, 4}
Compute R S, R S, R, S, R 1, S1 for R = {(1, 2), (2, 2), (3,
2)} and S = {(1, 3), (2, 3), (3, 3), (2, 1)} on set A = {1, 2, 3}.

4.4 CLOSURES
If R is some relation on A that doesnt have some
important relational properties like reflexivity, symmetry and
transitivity. Its natural to ask a question like can we make it
reflexive, symmetric and transitive by adding some pair in it and we

64

want to add as few new pairs as possible, so we want to find smallest

relation that contains given relation and posses the property we
desire. Such a smallest relation is called as closures of a given
relation.
4.4.1 Reflexive and Symmetric closures
Reflexive closure :Let R be a given relation on A. Then reflexive closure of
R is the smallest reflexive relation containing R and its denoted
by Rr .
Note : (1) If R itself is reflexive then R r = R
(2) R r = R
For example,
(1) Let A = {1, 2, 3} and R = {(1, 1), (1, 3), (2, 3), (3, 1)}

R r = (1,1), (1, 3), (2, 3), (3,1), (2, 2), (3, 3)

Symmetric closure :Let R be a given relation on set A. Then symmetric closure of

R is the smallest symmetric relation containing R and its
denoted by Rs .
Note : (1) If R itself is symmetric then R s = R
(2) R s = R R 1
For example,
(1) If A = { x , y, z} and R =

x, x , x, y , z, z , y, z
R s = x, x , x, y , y, x , z, z , y, z , z, y

Check your progress

1.

1
0
Let R be a relation whose matrix is
1

0 0 1
1 0 0
.
0 1 0

1 0 0 44

Find the reflexive and symmetric closures of R.

then

65

2.

Let R be a relation whose diagraph is

1
2
9

Fig. 4.1
Find the reflexive and symmetric closures of R.
4.4.2 Transitive Closure
Definition : Let R be a relation on A. The transitive closure of
R is the smallest transitive relation that contains R. Its denoted
by R T .
As compare to reflexive and symmetric closures its little bit
difficult to find transitive closure because we dont have a formula
for it but we have following results and an algorithm for finding the
transitive closure.
Result :(1) Let R be a relation on A. Then R is transitive closure of R.
For example,
(1) Let A = {1, 2, 3, 4} and R = {(1, 1), (1, 2), (2, 3), (3, 4)} we
know that if, |A| = m, then R = R1 R 2 .... R m .
2
1
3

4
Fig. 4.2

66

|A| = 4
R = R1 R 2 R 3 R 4
R1 = (1,1), (1, 2), (2, 3), (3, 4)
R 2 = (1, 2), (1, 3), (1,1), (2, 4)
R 3 = (1,1), (1, 2), (1, 4), (1, 3) and
R 4 = (1,1), (1, 2), (1, 4), (1, 3)

R = (1,1), (1, 2), (1, 3), (1, 4), (2, 4), (2, 3), (3, 4)
Transitive Closure = R T = R
The above graphical method is impractical for large sets and
relations and it is not systematic and also It would be more time
consuming and costly for large set.
But we have a more efficient algorithm for computing
transitive closure called as Warshalls Algorithm.
Warshalls Algorithm :Let R be a relation on a set A = a1, a 2 , ....., a n . If
x1, x2, x3 ,...., xn is a Path in R, then any vertices other than
x1 & xm are called interior vertices of the Path. Now, for 1 k n ,
we define a Boolean matrix Wk as follows. Wk has a 1 in position
i, j iff there is a path from ai to a j in R whose interior vertices, if
any, come from the set a1, a 2 ,....., a k .
So, it follows that Wn has a 1 in position i, j iff some Path
in R connects a i with a j , i.e. Wn = M R . If we define W0 to be
m R , then we will have a sequence W0 , W1,..., Wn = M R . Now, we

will see how to compute matrix Wk from the previous matrix Wk1
this procedure is called as Warshalls Algorithm.
Step I

Let W0 = M R .

Step II Suppose we have calculated Wk1 , now to calculate Wk .

67

Step III List the locations p1, p 2 ,... in column k of Wk1 , where
the entry is 1, and the locations q1, q 2 ,... in row k of
Wk1 , where the entry is 1.
Step IV Put 1s in all the positions pi ,q j of Wk (if they are not
already there.)
For example,
(1) Let A = {1, 2, 3, 4} and R = {(1, 1), (1, 2), (2, 3), (3, 4)} as
earlier.
Now, we will use Warshalls Algorithm to find R .
1
0
(1) W0 = M R =
0

1 0 0
0 1 0
0 0 1

0 0 0 44

(2) To computer W1 consider Ist column and Ist row of W1 where

1s are present.
1
0
W1 =
0

1 0 0
0 1 0
0 0 1

0 0 0

Column
1

Row
1
2

So fill the positions (1, 1) and (1, 2) by

1s if not present already.

(3) To compute W2 consider IInd column and IInd row of W1

where 1s are present.
1
0
W2 =
0

1 0 0
0 1 0
0 0 1

0 0 0

Column
1

Row
3

So fill (1, 3) by 1.

(4) To compute W3 , consider IIIrd column and IIIrd row of W2

where 1s are present.

68
1
0
W3 =
0

1 1 1
0 1 1
0 0 1

0 0 0

Column
1
2

Row
4

So fill (1, 4) and (2, 4) by 1s.

(5) To compute W4 , consider IVth column and IVth row of W3

where 1s are present.

W4 = W3 = M R

Column
1
2
3

Row
No
1s

So, no new entries.

R T = R = (1,1), (1, 2), (1, 3), (1, 4), (2, 3), (2, 4), (3, 4)
Note :
(1) Warshalls Algorithm gives another method to calculate R .
(2) Warshalls Algorithm relatively faster then other methods. (in
the sense of time taken)
Check your progress
1.

Let A = {1, 2, 3, 4} and matrix of Relation R; is,

1 0 1 0
0 1 0 0
.
MR =
0 0 1 0

1 0 0 0
Find R using (a) Warshalls Algorithm

(b) Diagraph
2.

Let A = { x , y, z, w} and matrix of Relation R is,

0
0
MR =
1

1 0 1
1 0 0
= W0
0 1 0

0 0 0

Compute W1, W2 , W3 as in Warshalls algorithm.

69

4.5 COMPOSITION :
Let A, B, C be given sets. Let R be a relation from A to B
and T be a relation from B to C then we can define a new
relation from A to C called as composition of R and T
[denoted by (TOR)]

ToR
T

R
A

Fig. 4.3
Definition : Let x A and y c, then x TOR y means there exist
w B such that ( x R w ) and (w T y).
For example,
(1) Let A = {1, 2, 3, 4} and
R = {(1, 1), (2, 3), (1, 4)} and T = (2, 2), (2, 3), (3, 4)}
(2, 3) R and (3, 4) T (2, 4) TOR

TOR = {(2, 4)}

In terms of Matrices,
If M R is a matrix related to R and
If M T is a matrix related to T and
M TOR is a matrix related to (TOR).

Then we have, M TOR = M R M T

For example,
(1) Let A = {1, 2, 3} and let R and T be relations on A whose
matrices
1 0 1
1 0 0

are M R = 1 1 1 and M T = 0 1 1 then we see that,

0 1 0
1 0 1

70

(1,1) R
(1, 3) R
(1, 3) R
(2,1) R
(2, 2) R
(2, 2) R
(2, 3) R
(2,3) R
(3, 2) R
(3, 2) R

and
and
and
and
and
and
and
and
and
and

(1,1) T (1,1) TOR

(3,1) T (1,1) TOR
(3, 3) T (1, 3) TOR
(1,1) T (2,1) TOR
(2, 2) T (2, 2) TOR
(2, 3) T (2, 3) TOR
(3,1) T (2,1) TOR
(3, 3) T (2, 3) TOR
(2, 2) T (3, 2) TOR
(2, 3) T (3, 3) TOR

1 0 1
M TOR = 1 1 1
0 1 1

We can check that, M TOR = M R M T so this formula gives an

easy way to compute ToR (by using M TOR )
Note : (1) If R = T then we have ToR = R 2 and
M TOR = M 2 = M R M R .
R

(2) If we have relation R, T, S such that,

R
A

T
B

Fig. 4.4
then To (SoR) = (ToS) o R [i.e. composition is associative]
(3) In General, RoS

SoR

(4) (SoR) 1 = R 1 o S1
Check your progress
1.

A = B = C , A = { x , y, z, w} and R and T be two relations

such that

71
1
0
MR =
0

0 0 1
1

1
1 0 0
and M T =
0
1 0 1

0 1 0
0

0 1 0
0 0 1
0 1 1

1 0 1

then compute (ToR) and (RoT)

2.

Give an example of relations R and T such that RoT ToR.

4.6 COMPUTER REPRESENTATION OF RELATIONS

AND DIAGRAPHS :
We know that a Relation R on set A can be represented by
an n X n matrix MR , if |A| = n. The matrix MR has entries that are
0 or 1. Then one of the easier way of representing R in a
computer is by an n X n array having 0s and 1s stored in each
location. Thus, if A = {1, 2} and R = {(1, 1), (2, 1), (2, 2)}, then
1 0
MR =
and these data would be represented by a two
1 1
dimensional array MR, where MR 1,1 = 1, MR 1, 2 = 0, MR 2,1 = 1,

MR [2, 2] = 1 (MR means matrix related to R)

An another way of storing data for relations and diagraphs is
by using the linked list idea of computer programming. A liked list
will be constructed that contains all the edges of the diagraph, that is,
the ordered pairs of numbers that determine those edges. The data
can be represented by two arrays, TAIL and HEAD, giving the
beginning vertex and end vertex, respectively for all edges.
If we are making these edge data into a linked list, then we
need an array NEXT of pointers from each edge to the Next edge.
Consider the relation whose diagraph is
5

2
2

7
4

5
10
9

Fig. 4.5

72

The vertices are the integers 1 to 6 and we arbitrarily

number the edges as shown in above diagraph. If we wish to store
the diagraph in liked list form so that the logical order coincides
with the numbering of edges. We can use a scheme mentioned
below.
START

TAIL

HEAD

NEXT

1
2
2
3
5
3
3
6
1
1

3
3
1
5
4
4
6
1
6
2

9
10
4
8
1
3
0
7
6
5

Fig. 4.6
START contains 2, the index of the first data item, the edge
(2, 3) [this edge is labeled with a 1 in fig : 5.6]. This edge is stored
in the second entries of TAIL and HEAD, respectively. Since
NEXT [2] contains 10, the next edge is the one located in position
10 of TAIL and HEAD, that is, (1, 2).
NEXT [10] contains 5, so we go to next to data position 5,
which contains the edge (5, 4). This process continues until we
reach edge (3, 6) in data position 7. This is the last edge, and this
fact is indicated by having NEXT [7] contains 0. We use 0 as a
pointer, indicating the absence of any more data.
If we trace through this process, we will see that we
encounter the edges in exactly the order corresponding to their
numbering. We can arrange, in a similar way, to pass through the
edges in any desired order.
But this scheme and the numerous equivalent variations of it
have important disadvantages. In many algorithms, its efficient to
locate a vertex and then immediately begin to investigate the edges
that begin or end with this vertex. This is not possible in general
with the mechanism shown in fig : 2. So we have modification of it.
We use an additional linear array VERT having one position for

73

each vertex in the diagraph. For each vertex I, VERT [I] is the
index, in TAIL and HEAD, of the first edge we wish to consider
leaving vertex I as shown below.
VERT

TAIL

HEAD

NEXT

10
2
4
0
5
8

1
2
2
3
5
3
3
6
1
1

2
3
1
5
4
4
6
1
6
3

0
3
0
6
0
7
0
0
1
9

Fig. 4.7
In Fig. 4.5, the first edge could be taken to edge with the
smallest number labeling it. Thus VERT, like NEXT, contains
pointers to edges. For each vertex I, we must arrange the pointers in
NEXT so that they link together all edges leaving I, starting with the
edge pointed to by VERT [I]. The last of these edges is made to
point to zero in each case. In a sense, the data arrays TAIL and
HEAD really contain several linked lists of edges, one for each
vertex.
In Fig. 4.7 we can see that VERT [1] contains 10, so the first
edge leaving vertex 1 must be stored in the tenth data position. This
is edge (1, 3). Since NEXT [10] = 9, the next edge leaving vertex
1 is (1, 6) located in data position 9. Again NEXT [9] = 1, which
points us to the edge (1, 2) in data position 1. Since NEXT [1] = 0,
we have come to the end of those edges that begin at vertex 1. The
order of the edges choosen here differs from the numbering in Fig. 1.
We then proceed to VERT [2] and get a pointer to position
2 in the data. This contains the first edge leaving vertex 2, and we
can follow the pointers to visit all edges coming from vertex 2. Note
that VERT [4] = 0, signifying that there are no edges beginning at
vertex 4.
So we have seen at least two methods for storing the data for
a relation or diagraph, one using the matrix of the relation and one

74

using linked lists. There are number of factors which determines the
choice of method to be used for storage and an analysis shows that
matrix storage method is n times faster than the linked list method
in most cases.

4.7 LET US SUM UP :

We saw the definition of an equivalence relation and method
to produce partition using an equivalence relation and vice versa.
Then we saw the different operations that can be performed on
relations and how some of them can be used to find closures. Then
we saw different method for computing transitive closures with an
important method Warshalls Algorithm. At the end we saw two
important concepts, composition of relations and computer
representation of relation and diagraph.

4.8 REFERENCES FOR FURTHER READING :

(1) Discrete structures by B. Kolman Hc Busby, S Ross PHI Pvt.
Ltd.
(2) Discrete mathematics and its Application Keneth H. Rosen
TMG.
(3) Discrete structures by Liu.

4.9 UNIT END EXERCISES

Q.1 Let A = , define the following relation R on A : (a, b) R
(c, d) iff a 2 + b 2 = c2 + d 2 show that R is an equivalence
relation.
Q.2 Let A = { x , y, z, w} and R be a relation on A defined by,
1
0
MR =
0

0 0 0
1 1 1
compute a partition produce by R.
1 1 1

1 1 1

Q.3 Let A = {1, 2, 3, 4, 12} and let R and S be the following

relations on A : x R y iff 2 | x y and x y iff 3 | x y .

75

Compute
(a) R

(f)

S1

(b) S

(g)

M R S

(c) R S

(h)

M R S

(d) R S

(i)

M 1
R

(e) R 1

(j)

MS

Q.4 Find the transitive closure of a relation whose matrix is

1
0

0
1

0 1 1 0
1 0 0 1
0 1 0 0

1 0 1 1
1 0 0 0

by (a) computing R (b) using Warshalls Algorithm

Q.5 Let A = {1, 2, 3, 4} and let R and S be the relations on A such
that
1
1
MR =
1

0 0 1
1

1
0 1 0
and MS =
1
1 0 0

0 0 1
0

0 1 0
1 0 0
0 0 1

0 1 1

use Warshalls Algorithm to compute the transitive closure of R S .

Q.6 If A = {1, 2, 3, 4} and let R and S be the relations on A such
that
1
1
MR =
1

1 0 0
1

0
0 1 0
and MS =
1
1 0 1

1 1 0
1

0 1 1
0 1 0
1 0 1

0 1 1

Find (a) M RoR , (b) M RoS , (c) SoR , (d) RoS .

76

5
PARTIAL ORDER SETS AND LATTICES
Unit Structure
5.0
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9

Objectives
Introduction
Definition and examples
Hasse diagrams
Isomorphism
External Elements of partially ordered sets
Lattices
Let us sum up
Unit End Exercise
References for further reading

5.0 OBJECTIVES:
After going through this chapter students will be able to understand:
The definition of partially order sets and example based on it.
Idea of Hasse diagram and able to represent the diagraph of a
poset in more efficient way.
The concept of Isomorphism and which is useful in
classification of Posets.
The concept of maximal, minimal elements, the greatest and
least element, upper and lower bound of a subset and finally
the concept of LUB and GLB.
The concept of Lattices and different properties of a Lattice.

5.1 INTRODUCTION:
We use relations to order some or all of the elements of sets.
For example we order words using the relation, containing pair of
words (x, y) where x comes before y in the dictionary. We
schedule projects using the relation consisting of pairs (x, y) where x
and y are tasks in project such that x must be completed before y

77

begins. When we add all the pairs of the form (x, x) to these
relations we get partial order. (Practical definition afterwards). These
structures are useful in set theory, algebra, sorting and searching, in
the construction of logical representations for computer circuits.

5.2 DEFINITION AND EXAMPLES:

Definition: Let R be a relation on set A. then R is said to be
partial order if R is (a) reflexive (b) antisymmetric and (c)
transitive.
The set A with partial order R is called as partial order set or
poset and Its denoted as (A, R)
For example
(1) Let A and R be a relation on A such that x R y iff x y.
Its easy to check R is reflexive, ant symmetric and transitive.
R is a partial order.
( , ) is a poset.

(2) Let S = 1, 2, 3and A = P(S)

A 0 ,1 , 2 , 3 , 1, 2 , 2, 3 , 1, 3 , S

Let R be a relation on A defined as x R w iff x w

(a) Let x A,
x x

xRx
xA
R is reflexive

(b) Let x, w A such that x R w and w R x

xRw x w

(1)

and wRx w x

(2)

w xfromand

R is antisymmetric

(c) Let x, w, z A such that x R w and w R z

xRw x w
(1)
and wRzw z

(2)

78

From and we have, x w z x z xRz

R is transitive
R is partial order
(A,R) is a poset

(3) Above example can be extended to any finite set as well as for an
infinite set
(4) Let A ,&'R ' be a relation on A such that x R y iff x
divides y.
Its easy to check (A, R) is a poset.
(5) Let A and R be a relation on A such that x R y iff x < y.
Then R isnt a partial order. ( ( Risn ' treflexive3 R 3)
(A, )isn ' taposet.

Note:
(1) Let R be a partial order then R 1 is also a partial order.
R 1 is called as the dual of R and the set (A, R 1 ) is called the

dual of the poset (A, R)

For example (a) dual of ( ,) is ( ,) .
(2) IF (A, R) is a poset, the elements x and y of A are said to be
comparable if x R y or y R x. The important observation is that
two elements in a poset may not be comparable.
For example 2 and 7 in with divisibility relation are not
comparable ( 27&72)
(3) If every pair of elements in a poset is comparable, we say that
A is a linearly ordered and the partial order is called a linear
order and we say A is a chain.
Check your progress
1. Determine whether relation R is a partial order on the set A.
(a) A ,andxRyiff x y is an even number.
(b) A ,andxRyiff x y

79

5.3 HASSE DIAGRAMS:

Result: The diagraph of a partial order has no cycle of length greater
than 1 (except at loops)
With the help of above result we can simplify the diagraph of a
partial order, simplified diagram is called as Hasse diagram.
Procedure for finding Hasse diagram:
1) Draw the diagram of given partial order so that all the arrows
are pointing in upward direction.
2) Remove all cycles of length 1 (i.e. loop at each of the
vertex)
3) Eliminate all edges that are implied by the transitive property
i.e. (a R b) and (b R c) then aRc so remove the edge from a
to c
c

Fig. 5.1
4) Replace arrows by line segments and circles by dots.
For example:
(1) Let A 1,2,4,5,10,20 with relation of divisibility.

80

20

10

Fig. 5.2
Diagraph
20

20

10

10

5
1

Fig. 5.3 Step: 1

Fig. 5.4

20

20

10

5
1

Step: 2

10

Fig. 5.5 Step: 3

Fig. 5.6 Step: 4

Hasse diagram

81

(2) Let S 1,2,3 andA P(S) 0,1 ,2 ,3 ,1,2 ,2,3 ,1,3S

with relation of contain
{ 1, 2, 3 } = S

{ 2, 3}
{ 1, 3}

{1,2}

{3}
{2}

{1}

Fig. 5.7 Hasse diagram

Check your progress:
Draw Hasse diagram of following diagraphs.
5

d
4

(a)
Fig. 5.8

(b)
Fig. 5.9
4

(c)
Fig. 5.10

82

5.4 ISOMORPHISM :
Let (A, R) and (B, T) be two Posets. Then (A, R) and (B, T)
are said to be Isomorphic if F a function f from A to B such
that
(a) f is bijective
(b) f preserves partial order. [i.e. for any a, b in A, we have
a R b iff f(a) T f(b)]
Note :- If (A, R) and (B, T) are Isomorphic then such a f is called
as an Isomophism from A to B.
For example,
(1) Let (A, R) = ,
(B, T) = (set of all even natural numbers, )
Define f : (A, R) (B, T) as f(m) = 2m
(a) Let a, b A such that f(a) = f(b)
2a = 2b
a=b
f is one one.
(b) Let b B

b = 2q q
Take a = q

f(a) = f(q) = 2q = b
f is onto.
f is bijective.

true b B

Now, Let a, b A such that a R b a b

2a 2b
f(a) f(b)

f preserves partial order.

f is an Isomorphism.
(A, R) and (B, T) are Isomorphic.

83

5.5

EXTERNAL
ELEMENTS
ORDERED SETS :

OF

PARTIALLY

Let (A, R) be a given poset.

An element a A is called a maximal element of A if there is
no element x (except a) in A such that a R x .
Similarly, an element b A is called a minimal element of A
if there is no element x (except b) in A such that x R b .
For example,
(1) Consider the Hasse diagrams of some posets as shown below.
0

6
5
3

(a)

8
2

(b)

Fig. 5.11
Fig. 5.12
Maximal element = 0
Maximal elements = 5, 6
Minimal element = 1
Minimal elements = 0, 1, 2
(2) Consider a poset , , minimal element = 1, maximal element
doesnt exist.
(3) Consider a poset , minimal and maximal doesnt exist.
Result :Let A be a finite non empty poset, then A has atleast one maximal
and atleast one minimal element.
An element a A is called a greatest element of A if x R a
x A and an element b A is called a least element of A if bRx
x A .

84

For example,
(1)
3

Greatest element = 3
Least element = 1

Fig. 5.13

(2) Poset , greatest element doesnt exist least element = 1

(3)
e

Fig. 5.14
Greatest element = e
Least element doesnt exist
(4)
d
e

Fig. 11.15
Greatest as well as least element doesnt exist.

85

Result :A poset has at most one greatest and at most one least element.
Note :
(1) Greatest element is denoted by I and its also called as unit
element
(2) Least element is denoted by 0 and its also called as zero
element.
Consider a poset (A, R) and a subset B of A. An element
u of A is said to be an upper bound of B if bRu b B .
An element l of A is said to be a lower bound of B if
bB.
For example,
lRb

(1) Consider the following poset and find the upper and lower
bounds of (a) B1 = {2, 5, 7} (b) B2 = {1, 3} (c) B3 = {3, 2, 7, 4}
3

1
2

7
6

4
5

Fig.5.16
For B1 ,

Upper bounds are 1, 2, 3

Lower bound is 5

For B2 ,

No upper bounds
Lower bounds are 2, 7, 4, 5, 6

For B3 ,

Upper bound is 3
Lower bounds are 5, 6

Note : 1) A subset B of a poset may or maynt have upper or

lower bounds.
2) It can have more than one upper or lower bounds.

86

3) Upper or lower bound may or maynt belong to given

subset.

1)

Let A be a poset and B a subset of A.

An element are u A is called a least upper bound of B,
L B B

(a)
(b)

If u is an upper bound of B.
If w is any other upper bound of B then u R w,

Similarly,
2)
An element l A is called a greatest lower bound of B,
GLB B

(a) If l is a lower bound of B.

(b) If t is any other lower bound of B then t R l.
For example,
(1) Consider the following poset
(a) for a subset B = {e, d, h}
Upper bounds = a, c, b
Lower bounds = h, f, g
L B B = h

GLB B = does not exist (as a and c are not comparable)

b

d
h
f

Fig. 5.17
(b) For a subset D = {a, d. f. g}
Upper bounds = a, b L B = a
No lower bounds GLB = does not exist

87

(c) For a subset E = {a, e, d}

Upper bounds = a, b L B = a
Lower bounds = h, f, g
GLB = h
Result :
Suppose that (A, R) and (B, T) are Isomorphic posets under the
Isomorphism f : A B
(a) If a is a maximal (minimal) element of (A, R), then f(a) is a
maximal (minimal) element of (B, T).
(b) If a is the greatest (least) element of (A, R), then f(a) is the
greatest (least) element of (B, T)
(c) If a is an upper bound (lower bound, least upper, upper bound,
greatest lower bound) of a subset D of A, then f(a) is an upper
bound (lower bound, least upper bound, greatest lower bound)
for the subset f(D) of B.
[Notation : if D = {1, 2, 3} then f(D) = {f(1), f(2), f(3)}]
(d) If every subset of (A, R) has a L B (GLB), then every subset of
(B, T) has a L B (GLB).
For example,
(1)
d

Fig. 5.18

Fig. 5.19

This two posets are not Isomorphic as first one have a greatest
element but second one not.
(2) Consider the posets, A = {1, 2, 3, 6} with divisibility relation and
S = {1, 2} , A = P(S) with contain relation

88
6

{1}

{2}

Fig. 5.20

Fig. 5.21

So these two posets are Isomorphic under the Isomorphism f

which is defined as, f(1) = , f(3) = {2}, f(2) = {1}, f(4) = S and
they satisfy all conditions stated in above result.
Check your progress
1. Consider the poset,
m
b

t
e

h
f

(a)

Fig. 5.22
Find maximal, minimal, greatest and least element if exist.
(b) For following subsets find
(i) All upper bounds
(ii) All lower bounds
(iii) The least upper bound
(iv) The greatest lower bound
1) B1 = {a, c ,h}
2) B2 = {h, f, g, e}
3) B3 = {a, d, e, f}
4) B4 = {b, c, h, e}
2. Check whether following posets are Isomorphic if yes find the
corresponding Isomorphism.

89
b

t
d
f

Fig. 5.23

Fig. 5.24
1

4
2

II

1
5

Fig. 5.25

Fig. 5.26
6

III

Fig. 5.27

Fig. 5.28

3. Find maximal, minimal, greatest and least element if exist.

(a) A = with usual partial order .
(b) A = { x / x R and 0 x 1} with usual partial order
(c) A = { x / x R and 0 x 1} with usual partial order
(d) A = { x / x R and 0 x 1} with usual partial order

90

5.6 LATTICES
5.6.1 Definition
A lattice is a poset (L, R) in which every subset consisting of
two elements has a least upper bound L B and greatest lower
bound (GLB).
Notation :- If a set is {a, b} then L B({a, b}) is denoted by a b and
its called as join of a and b. Similarly, GLB ({a, b}) is denoted
by a b and its called as meet of a and b.
d

For example
(1) Consider a poset,

c
b
a

Fig. 5.29

a
b
c
d

a
a
b
c
d

b
b
b
d
d

c
c
d
c
d

d
d
d
d
d

a
b
c
d

a
a
a
a
a

b
a
b
a
b

c
a
a
c
c

d
a
b
c
d

So, from the table of join and meet we can see L B and GLB of
any subset with two elements exist.
its a Lattice.
(2) Consider a poset,

Fig. 5.30
Since L B ({ x, y}) doesnt exist.
its not a lattice.

91

(3) Let S be a given set and L = P(S). Consider L with relation

contain
We know that ( L, ) is a poset.
Let x, y L, then x y = X Y and X Y = X Y
and which exist in L. (as they are subsets of s)

(L, ) is a lattice.

(4) Consider the poset ( , divisibility relation),

Let x, y ,

then x y=LCM(x, y) and x y=GCD(x, y)

(5) Let n ,
Let Dn = set of all positive divisors of n, then it can be proved
Dn with divisibility relation is a poset.
For example,
If n = 20, then D20 = {1, 2, 4, 5, 10, 20}
20

10

Fig. 5.31
Hasse diagram of D20
is a Lattice. [it can be proved similarly as in example (1).]

92

Check your progress

1. Check whether following Hasse diagrams are lattice or not?
f

e
e

c
b

Fig. 5.32

Fig. 5.33

(a)

(b)
t

2
y

Fig. 5.34
(c)
5.6.2 Isomorphic Lattices

L1,

R1 and

L2 ,

w
z

Let

Fig. 5.35
(d)

R 2 be two given posets.

A function

f : L1 L 2 is said to be an isomorphism if

(1) f is bijective
(2) f preserves the Lattice properties. [i.e. if a, b L1 then
f a b = f a f b and f a b = f a f b ]
Note : L1, R1 and L2 , R 2 are said to be Isomorphic Lattices.
For example,
(1) ( D6 , divisibility) and P 1, 2, 3 , are Isomorphic Lattices.

93

5.6.3 Properties of Lattices

Theorem :Let (L, R) be a lattice, then for every a and b, c in L,
(1) a b = b iff a R b
(2) a b = a iff a R b
(3) a b = a iff a b = b
(4) a a = a
(5) a a = a

Indempotent Properties

(6) a b = b a
(7) a b=b a

Commutative Properties

(8) a b c = a b c
(9) a b c = a b c

Associative Properties

(10) a a b = a
(11) a a b = a

Absorption Properties

Proof :(1)

Assume a b=b

a R (a b )
a R b

Conversely, if a R b, since b R b, b is an upper bound of a

and b. So, by definition of least upper bound we have
a b R b . Since, a b is an upper bound, b R a b , so
ab b.
(2)

and (3) proof similar to (1).

(4)

and (5) follows from definition of L B and GLB.

(6)

a b = L B ({a, b})

= L B ({b, a})
= ba
(7)

Proof similar to (6)

94

(8)

From definition of L B , we have a R a b c and

b c R a b c and also b R b c and c R b c so by

transitivity, b R a b c and c R a b c thus,
a b c is an upper bound of a and b, so by definition of
L B we have, a b R a b c .
Since, a b c is an upper bound of a b and c, we obtain

a b c R a b c

Similarly, a b c R a b c
By antisymmetry of R, we have a b c = a b c
(9)

Proof Similar to (7)

(10) Since, a b R a and a R a, we have a is an upper bound of

a b and a so a a b R a .

On the other hand, by definition of L B , we have

a R a a b , so a a b =a
(11) Proof is similar to (10).

5.7 LET US SUM UP

We started the concept of partial order set (poset), a set with
some special properties of a relation, defined on a set. Then we saw
if a given relation is a partial order then diagraph of a partial order
can be represented in better way, so that we can retrieve more and
more properties of a given poset. Then we had seen a concept of
Isomorphism, which divides the space of Lattices in different
groups. Then we saw the concept of maximal, minimal elements,
upper bounds, lower bounds, GLB and L B of a subset of a given
poset. GLB and L B concepts are useful in defining Lattice which
in turn useful in defining Boolean Algebra (which we will be seeing
in the next chapter).

95

5.8 UNIT END EXERCISES

1. Determine whether the following relation R is a partial order or
not on a given set.
(a) A = , a R b iff a + b is even.
(b) A = , a R b iff a = b.

2. Determine the Hasse diagram of the relation on A {1, 2, 3, 4, 5}

whose matrix is

(a)

1
0

0
0

0 1 1 1
1 1 1 1
0 1 1 1

0 0 1 0
0 0 0 1

1
1

(b)

0 0 0
1 0 0
1 1 0

1 0 1

3. Draw the Hasse diagram of Poset with partial order divisibility)

and determine which posets are linearly ordered.
(a) A = {1, 2, 3, 4, 6, 9, 12, 36}
(b) A = {3, 6, 12, 36, 72}
4. A is the set of all 2 X 2 Boolean matrices and the relation R is
defined as M R N iff mij nij , 1 i 2 , 1 j 2 .
(a) Find maximal and minimal elements of A
(b) Find the greatest and least element if exist of A.
(c) Find

all

upper

and

lower

bounds

of

1 0 1 0 1 0
B=

0 1 , 1 1 , 0 0

(d) Find GLB and L B of above set B.

5. Draw the Hasse diagram of D30 , D36 , D42 .
6. Determine whether following Hasse diagram represents a Lattice
or not.

96
d

(i)

(ii)

5
2

f
1

Fig. 5.36

Fig. 5.37
n

a
b

m
w

(iv)

(iii)

Fig. 5.38

Fig. 5.39

5.9 REFERENCES FOR FURTHER READING

(1) Discrete structure by B. Kolman, Hc Busby, S. Ross PHI Pvt.
Ltd.
(2) Discrete structures by Liu.
(3) Discrete mathematics and its Apprlications Keneth H. Rosen
TMG.

97

6
FUNCTION
Unit Structure
6.0 Objectives
6.1 Introduction
6.2 Functions
6.3 Types of function
6.4 Identity functions
6.5 Composite function
6.6 Inverse function
6.7 Binary operation
6.8 Properties of binary operation
6.9 Review
6.10 Unit End Exercise

6.0 OBJECTIVES:
A function is the central to the study of physics and
enumeration.
In computer implementation of any program output of any
program can be considered as a function of the input.
Binary operations have applications in the study algebraic
structures.

6.1 INTRODUCTION:
A function was the heart of the scientific revolution of the
seventeenth century. To understand the general use of function we
must study their properties in the general, which is what we do in
this chapter.
The reader is no doubt familiar with function of the form
y f ( x) for instance, if f ( x) x 2 2 , x 2 and y f (2) then the
value of y is 4.

98

6.2 FUNCTIONS:
Definition: Let X and Y are non-empty sets. A function f form
X to Y if for each element of a X exactly one element b Y (called
the image of a under f ) such that a is in relation f to b, then is
called function or mapping from X to Y.
If f is a function (mapping) from X to Y we write
f
f : X Y Or X
Y .
Suppose f is a any function from X to Y. Here set X is called
domain of the function f , and set Y is called co-domain of
function f . The image of a function f : X Y is the
set f ( X ) f (a ) / a X . The pre-image or inverse image of the
function f : X Y is the set f 1 (b) a / f (a ) Y .
The range of a function f : X Y as the image of its domain. i.e. set
f (X ) .
For the function y f ( x ) , y is also known as the output
corresponding to the input x.
Note: Every element a X has an image, but it is not necessary that
every element b Y also has pre-image in X.
Example-1. State whether the following are the function or not. If
X p, q, r , s and Y 1, 2,3, 4 . Give reasons and also find the range
of the function.
(i) f p,1 , q, 2 , r , 2 .

(ii) g p,1 , q,3 , r , 2 , q, 4 .

(iii) h p,1 , q,3 , r , 2 , s, 4 .

Solution.
(i) Since the element s X do not have image in Y, therefore f
is not a function from X to Y.
(ii) The element q X it has two different images in Y i.e. 3 and
4 Y . Therefore g is not a function from X to Y.
(iii)
By definition of function, each element of X has exactly
one image in Y. Therefore h is a function from X toY. So
range of h h X 1, 2,3, 4 Y .

99

Example-2. Let the function

be
f :X
3
by f ( x) x 3x 2 . If X 1, 0,1, 2 then find range of f .

defined

Solution.
Given that f is the function from X . By definition of range we
have to find image of each element of X.
f (1) 1 3 1 2 2
3

f (0) 0 3 0 2 2
3

f (1) 1 3 1 2 6
3

f (2) 2 3 2 2 16
3

Thus the range of f is the set 2, 2, 6,16 .

6.3 TYPES OF FUNCTION:

Definition: A function f : X Y is
said to be injective or
(oneone or monomorphism) if x1 , x2 X : f ( x1 ) f ( x2 ) x1 x2 or
equivalently f ( x1 ) f ( x2 ) x1 x2 .
Definition: A function f : X Y is said to be surjective or
(Onto, or epimorphism), if every y Y such that y f ( x )
for some x X .
A function that is injective and surjective is said to be bijective, if
f : X Y is a bijective function we may write f : X Y . For e.g.
Let the function f : X Y is defined by
a

Y
Fig.6.1

Here f (a ) 1, f (b) 2, f (c) 3 .

f ( x1 ) f ( x2 ) x1 x2
f is injective.
y Y x X So that f ( x ) y .
f is surjective.

(I)
( II )

100

from I and II we say that

f is bijective.
e.g.
If f : A B ,
defined
by f (1) a, f (2) a, f (3) b .
Where A 1, 2,3 , B a, b, c . Then check whether the function is
bijective or not?
Solution: If f : A B , defined
Where A 1, 2,3 , B a, b, c .

by f (1) a, f (2) a, f (3) b .

Here, f (1) f (2) but 1 2 .

f : A B is not injective.
f : A B is not bijective.

.
Example-3 If function f : is defined by f ( x) 3x 1
then prove that f is bijective.
Solution. We first show that f is injective i.e.
f ( x1 ) f ( x2 ) x1 x2

So we assume that
f ( x1 ) f ( x2 )
3 x1 1 3x2 1
3 x1 3 x2

(Adding 1 both side)

x1 x2
(Dividing both side by 3)
(I)
Thus f is injective.
Now to show that f is surjective.
Let y be a real number. We must find value of x such that f ( x) y .
Thus we must find x such that y 3x 1 solving,
x

y 1

101
Thus f is surjective.

(II)

from I and II we say that

f is bijective.

6.4 IDENTITY FUNCTIONS

Definition: A function f on X is said to be identity function if
f ( x) x for every x X . It is denoted by I x and leaves every input
unchanged. For e.g. Let X 1, 2,3 and f 1,1 , 2, 2 , 3,3
which can be written in equation f (1) 1 , f (2) 2 , f (3) 3 is an
identity function of X .
Theorem 6.1. Let f : X Y for each subset A Y ,
f ( f 1 ( A)) A .
Proof. Let b f ( f 1 ( A))
f (a ) b for some a f 1 ( A)
b A
f ( f 1 ( A)) A .

Theorem 6.2.If f is function form X onto Y then f ( f 1 ( A)) A .

Proof. Above theorem 4.1 we prove that f ( f 1 ( A)) A .
For opposite inclusion let b A then there exist some element
a X such that f (a ) b because f onto.
f (a ) b A a f 1 ( A)
f (a) f ( f 1 (a ))

b f ( f 1 ( A))
A f ( f 1 ( A))

Hence f ( f 1 ( A)) A .
Theorem 6.3. Let function f : X Y for each subset A X ;
Then A f 1 ( f ( A)) .
Proof: Let f : X Y for each subset A X .
Let a A f (a ) f ( A)
(By definition of pre-image)
a f 1 ( f ( A))
A f 1 ( f ( A)) .
Hence A f 1 ( f ( A)) .

102

Theorem.6.4. If f is one-to-one function from X onto Y, then

A f ( f 1 ( A)) .
Proof. In above theorem we prove that A f 1 ( f ( A)) .
Let a f ( f 1 ( A)) f (a ) f ( A)
f (a ) f (a ' ) For some a ' A .
a a ' Since f is one-to-one.
a A
f ( f 1 ( A)) A .

Hence A f ( f 1 ( A)) .

6.5 COMPOSITE FUNCTION

Definition: Let f : X Y and g : Y Z then the composite of the
function f and g denoted by ( gof ) is a function of X Z given by
( gof ) : X Z such that ( gof )( x) g f ( x) , x X .
For e.g. Let X p, q, r , Y 1, 2,3 and Z c, d , e . Let f : X Y
and
be
defined
by
g :Y Z
f p, 2 , q,1 , r ,3 and g 1, d , 2, c , 3, e .
Then
the
composite function ( gof ) : X Z may be computed in the following
manner: ( gof )( p ) g f ( p) g (2) c ,
( gof )(q ) g f (q ) g (1) d , ( gof )(r ) g f (r ) g (3) e .

Thus ( gof ) : X Z is given by ( gof ) p, c , q, d , r , e .

Example-4: Let f : is defined by f ( x) x 1 , x ,
And the function g : is defined by g ( x) x 2 , x .
Find gof and fog .
Solution: Let the composite function gof : is given by
( gof )( x) g f ( x) g ( x 1) ( x 1) 2 . And the composite function
fog : is given by ( fog )( x) f g ( x) f ( x 2 ) x 2 1 .
gof fog

Remark; In general the composite function is not commutative.

Theorem 6.5. If f : W X , g : X Y and h : Y Z , then
( hog ) of ho ( gof ) .
Proof. It is clear that (hog )of and ho( gof ) is the function from W
to Z. This two function will be equal if they have the same image to
each element x W .

103

Now, ((hog )of )( x) (hog )( f ( x))

h( g ( f ( x )))
h( gof ( x))
(ho( gof ))( x)
Which shows that ( hog ) of ho ( gof ) .

6.6 INVERSE FUNCTION

Definition: Let f : X Y be a one-to-one function. The inverse of
function denoted by f 1 , is the set of f 1 y, x / x, y f .
This, if
is a one-to-one function, then
f : X Y
1
f ( x) y x f ( y ) for x X and y Y .
Theorem 6.6. The inverse of a one-to-one and onto function is
unique.
Let
f : X Y be one-to-one and onto function. Let
g : Y X and h : Y X be two different inverse function of f .
Then for x1 , x2 X there exists y Y such that g ( y ) x1 , h( y ) x2 .
g ( y ) x1 y f ( x1 )
g is the inverse of f .
Proof.

h( y ) x2 y f ( x2 )

h is the inverse of f .

Then it follows that f ( x1 ) f ( x2 ) x1 x2 . f is one-to-one.

g ( y ) h( y ) For y Y .
Hence the inverse of a one-to-one and onto function is unique.
Theorem 6.7. The inverse of a one to one function is one to one and
onto.
Proof. Let f : X Y be one-to-one and onto function.
Then for x1 , x2 X there exists y1 , y2 Y such that f ( x1 ) y and
f ( x2 ) y2 x1 f 1 ( y1 ) and x2 f 1 ( y2 ) .
Now for y1 , y2 Y ,
f 1 ( y1 ) f 1 ( y2 ) x1 x2
f ( x1 ) f ( x2 )
y1 y2

This proves that f

is one-to-one.

Again since f is onto, then for y Y there exists some x X such

that f ( x) y x f 1 ( y )
f is one-to-one
1
This proves f is onto.

104

Example-5 If function f : be defined by f ( x) 2 x 3 for

every x , has its inverse ( f is invertible). Find formula for f 1 .
Solution: Let f : be defined by f ( x) 2 x 3 for if x1 , x2
then f ( x1 ) f ( x2 ) 2 x1 3 2 x2 3
2 x1 2 x2
(Adding -3 both side)
x1 x2
(Divide by 2 both side)
This proves that f is one-to-one. ( f is injective)
I
Again, if y , y f ( x) y 2 x 3
x

y 3
2

1
( y 3) such that
2
1

f ( y 3) 2 ( y 3) 3
2

y 3 3 y .
This proves that f is onto. ( f is surjective)
From I and II we say that f is bijective.

Thus for x there exists

II

f 1 is exists and it is defined by

1
f 1 ( y ) ( y 3) .
2

Example.6. If each functions f and g is one-to-one then function

gof is one-to-one.
Solution: Let f : X Y and g : Y Z is one-to-one function.
Let x1 , x2 X then gof ( x1 ) gof ( x2 )
g ( f ( x1 )) g ( f ( x2 ))
f ( x1 ) f ( x2 )
x1 x2

(g is one-to-one)

( f is one-to-one)

Hence function gof is one-to-one.

Examlpe.7. If each function f and g is onto then function gof is
onto.
Solution: Let f : X Y and g : Y Z is onto function.
By definition of composite function gof : X Z .To prove that gof
is onto we have to prove that every element z Z is an image
element for some x X under gof . Since g is onto y Y such
that g ( y ) z . Again f is onto x X such that f ( x) y .
Now, gof ( x ) g ( f ( x)) g ( y ) z .
Hence function gof is onto.

105

Example.8. Let A and B be two non-empty sets. Let f : A B be

the function then prove that, If A1 and A2 subset of A so that A1 A2
then f ( A1 ) f ( A2 ) . Is the converse true?
Solution: Let y f ( A1 ) x A1 such that y f ( x) .
Since A1 A2 , x A2 such that y f ( x) .
f ( x) f ( A2 ) y f ( A2 )
f ( A1 ) f ( A2 ) .

But the converse is not true.

To show that the converse is not true we give counter example.
Let f : A B be the function. f (a) 1 , f (b) 1 , f (c) 2
A a, b, c , B 1, 2
Let A1 a, c and A2 a, b be the subset of A .
f ( A1 ) 1, 2 and f ( A2 ) 1, 2 .
f ( A1 ) f ( A2 )

f ( A1 ) f ( A2 ) But A1 A2 .

EXERCISE 6.1
1. Decide whether or not the following are functions from A to B
where A a, b, c, d , e and B p, q, r , s . If they are function, give
the range of each. If they are not tell, why?
(i) f a, p , c, q , e, s , d , r .
(ii) g a, t , e, s , c, p , b, r a, s , d , p .
(iii) h a, p , b, s , c, t , d , q , e, r .

2. Each of the following formulas defines a function from to .

Find the range of each of the function.
(i) f ( x) x3 .
(ii) g ( x) sin x .
(iii) h( x) x 2 1 .
3. Let be the set of rational numbers. Let f : be defined
by f ( x) 2 x 3 , x . Show that f is bijective. Also find a
formula that the inverse function f 1 .
4. Prove the following two result for a finite set T :
(i) If f maps T onto T then f is one-to-one.
(ii) If f is a one- to-one mapping of T into itself, then f is onto.

106

5. Let X 3 and Y 1 where is the set of real

numbers. Let the function f : X Y be defined by f ( x)

x2
is
x 3

this function is bijective?

6. Prove that identity function of set T into itself is one-to-one and
onto.
7. If f : X Y and g : Y Z be two one-to-one and onto function,
then gof : X Z is also one-to-one and onto also Show
that ( gof ) 1 : Z X ( f 1og 1 ) Z X .

6.7 BINARY OPERATION

The basic idea underlying the definition of an algebraic
structure is that of a set with a binary operation. Suppose we have a
set A of objects with the property that any pair of them x and y , can
be combined in some way to form an objects z . This can be
expressed by the equation
x y z

Where the indicates a binary operation. The word binary

signifying here that two objects are involved. The most familiar
examples are the arithmetical operation like + and defined on the
set of integers .
Definition: Given a set of element A , then a binary operation on
the set A is a rule of combination which assigns to each ordered
pair of element
a, b A a unique element c A . We write
symbolically that c a b .
Example.9. Let be the set of all natural numbers then the
operation of addition on the set is a binary operation for if
a, b then so is c where c a b .
Example.10. Let A be the of all odd integers then the operation of
addition on the set A is not a binary operation for if a, b A then
c A where c a b .
Remark: Addition, Multiplication are the binary operation
in , , , .
Subtraction is binary operation in , . Division is not a binary
operation.

107

6.8 PROPERTIES OF BINARY OPERATION

Commutative: A binary operation * on a set of element A is said to
be commutative, if and only if, for a, b A ,
a *b b * a

Associative: A binary operation * on a set of element A is said to be

associative, if and only if, for every a, b, c A ,
a *(b * c ) (a * b) * c

Distributive: A binary operation * on a set of element A is said to

be distributive over the binary operation on the same set A
elements if and only if for every a, b, c A ,
a *(b c ) (a * b) (a * c)

Identity element: An element e in a set A is said to be a identity

element with respect to the binary operation * on A if and only if for
every a A ,
a *e e*a a

Inverse element: An element b is in a set A is said to be inverse

element of an element a A with respect to the binary operation *,
if and only if,
a *b b * a e

Example.11. Check the following operation is commutative and

associative. a * b a b 10 , For every a, b .
Solution: For every a, b .
(i) a * b a b 10
b a 10
b*a
* is a commutative.

(ii) a *(b * c) a *(b c 10)

a b c 10 10
a b c 20

(a * b) * c (a b 10) * c
a b 10 c 10
a b c 20

II

From I and II we say that,

a *(b * c) (a * b) * c
* is an associative.

108

Example.12. The binary operation * on the set of all real numbers

are defined by a * b a b . Show that the * is commutative but not
associative.
Solution: Since a * b a b
ba
b*a
* is a commutative.

Again,
( a * b) * c a b * c

a b c
a *(b * c) a * b c
a bc

If a 1, b 2, c 5 then
( a * b) * c 1 2 5 1 5 4
a *(b * c ) 1 2 5 1 3 2

* is not associative.

Example.13.

binary

operation

0 as a * b

ab
. a, b 0
9

is

defined

on

Show that the * is closed under commutative, associative and also

find identity and inverse element of the binary operation.
Solution: For a, b, c 0 ,
(i) a * b

ab ba

b*a
9
9

* is a commutative.
ab
c abc
9
ab
(ii) (a * b)* c * c
9
81
9
bc
a
abc
9
bc
a *(b * c) a *
9
81
9
(a * b) * c a *(b * c)
* is associative.

109

(iii) Let e be an identity element in 0 with respect to *.

a*e a
a *e

a
9
e 9

9 0 , a 0

e 9 be an identity element in 0 with respect to *.

(iv) Let b be an inverse element in 0 with respect to *.

a *b e
a *b 9
e 9
ab
9
9
81
b
a
81
be an inverse element in 0 with respect to *.
b
a

Exercise 6.2
1.
Check whether the following binary operation defined on
corresponding set is commutative or associative.
(i) a * b 2a b
for a, b
(ii) a * b a b 3
for a, b
(iii) a * b 2a 3b
for a, b
(iv) a * b a 2b

for a, b

2.

Find identity and inverse of element in .

Where a * b a b 10 , a, b .

3.

A binary operation * is defined as a * b

a 2b 2
, a, b 0 ,
4

Show that * is commutative and associative also find identity and

inverse element in 0 with respect to *.

6.9 REVIEW
In this chapter we have covered the following points:

Function and term related to function ( range, image, Preimage).

Injective function, surjective function and bijective function.
Identity function, composite function, inverse function.
Binary operation and their property.

110

6.10 UNIT END EXERCISE

1. Define the term give example of each term.: function, range of
function, Image of function, pre-image of the function.
2. Explain all the of function with their definition and counter
example.
3. Decide whether or not the following are functions from A to B
where A p, q, r , s, t , u and B a, b, c, d . If they are function, give
the range of each. If they are not tell, why?
(i). f p, a , q, b , s, c , r , d
(ii) g t , a , s, a , p, c , r , b s, d .
(iii) h p, a , s, b , t , d , q, c .

4. Each of the following formulas defines a function from to .

Find the range of each of the function.
(i) f ( x) x3 2 .
(ii) g ( x) cos .
(iii) h( x) x3 5 x 6 .
5. Find the image set of the function

f : defined by

3x
.
x 1
6. Let be the set of rational numbers. Let f : be defined
by f ( x) 9 x 4 , x . Show that f is bijective. Also find a
f ( x)

formula that the inverses function f 1 .

5
7. Let X and Y 0 where is the set of real
2

numbers. Let the function f : X Y be defined by f ( x)

1
is
2x 5

this function is bijective?

8. Show that f : is given by f ( x) x are neither injective
nor surjective.
9. Let the functions f : , f ( x)

2
, g : , g ( x) x 2 3
x 1

and h : , h( x) 3 x 2 then determine the following composite

function (i) hog , (ii) fog , (iii) gof , (iv) foh , (v) hof , (vi) goh .

111

f : and g : be two function given by

f ( x) x x for all x and g ( x) x x for all x . Find fog

10. Let

and gof .
11. Let f : A B be a function. If B2 B1 B , then show
that f 1 ( B2 ) f 1 ( B1 ) .
12. Check whether the following binary operation defined on
corresponding set is commutative or associative.
(i) a * b a bb a
for a, b
(ii) a * b 2a 2b 8
for a, b
ab
(iii) a * b 2
for a, b
(iv) a * b ab a b
for a, b
13. Find identity and inverse of element in .Where a * b a b 3 ,
a, b .

112

7
PIGEONHOLE PRINCIPLES
Unit structure:
7.0
7.1
7.2
7.3
7.4
7.5

Objectives
Introduction
Pigeonhole principle
The extended pigeonhole principle
Let us sum up
Unit End Exercise

7.0 OBJECTIVE
After going through this chapter you will be able to:
learn different counting techniques like Pigeonhole Principle
define cardinality of sets
learn about the properties related to cardinality of sets

7.1 INTRODUCTION
This chapter dedicated to the study of size of sets and also
we show how one can efficiently do the counting in a variety of
situations. We represent the basic principle of counting which is
easily derived and extremely useful. We know that some sets are not
finite there for Cantor gives cardinality of infinite sets in 1870s and
1881s.
Suppose there are 7 peoples working in an office and a pile
of 8 letters is delivered. Each letter is addressed to are of the people
in the office, and letter are put in the appropriate pigeonholes. Then
we know for sure that some are is going to be lucky and get more
then one letter. Generally it seen obvious that if n m and n
letters are put into m pigeonholes then one pigeonhole will
receive more then one letter.

113

If the statement is obvious then so also must be its

cantrapositive, because that is logically equivalent to it. The
cantrapositive is: if every pigeonhole receives at most one letter,
then n m .
Now we can formulate a purely mathematical form of this
principle we suppose that the letters are numbered
1, 2,3, 4 , n and the pigeonholes are numbered 1, 2,3, 4 , m for
each letter the address tells us which pigeonhole should receive it.
Thus in mathematical terms we have a rule for assigning letters to
pigeonholes that is a function from set n to the set m . The
condition that every pigeonhole receives at most one letter is
equivalent to the condition that this function is an injection.
1

-------------

n-1

m-1

Putting letter into pigeonholes

1

--------------

Thus the cantra positive from the original statements

essentially: if a function n m is an injection.
Then it must follow that n m . Now we know that what we
have to prove.
Theorem : 7.1 Let m be a natural number then the following
statements true for every natural number n if there is an injection
from n to the set m then n m .
Proof : We use the principle of induction. The statement is true
when n 1 , since 1 m for any natural number m . The induction
hypothesis is that the statement is true when n taking a specific
value k 1 . We have to deduce that it is true when n k 1 .
Suppose that f : n m is an injection. Since k 1 2 it
follows that m can not be 1. So m s 1 for some natural number s .
In order to show that k 1 m s 1 . We construct an injection
f : k s and uses the induction hypothesis conclude that k s .
There are two cases:

114

(I) Case:1: Suppose that f ( x) s 1 for all x k . Then let

be
*
the injection defined by f x f x for all x k .
(II) Case:2: Suppose that there is an x k such that f ( x) s 1 .
Then f (k 1) y . Where (since f is injection) y s 1 in this case
defined f * as follows.
f * x y ,

f * z f ( z ) ( z x) .

It is easy to check f * is an injection n to m .

1

1
f*

k+1

s+1

The theorem is contrapositive from the original statement:

If n m then there is no injection from n m

7.2 PIGEONHOLE PRINCIPLE

We represent the basic principle of counting which is easily
derived and extremely useful.
Statement: If there n -pigeons to be placed in m -pigeonhole
where m n . Then there is at least one pigeonhole which receives
more then one pigeon.

Pigeonhole Principle
Here is a simple consequence of the pigeonhole principle.

115

In one set 13 or more people there are at least two whose

birthdays fall in the same month.
In this case we have to think of putting the people in to
pigeonhole. it can be January, February, March and so on. Since
there are 13 people and only 12 pigeon holes one of the pigeonhole
must contain at least two people.
That this intuitively obvious result can be quite useful is
illustrated by the following example.
Example 1:
If eight people are chosen in any way what so ever at least
two of them will have been born on the same day of the week.
Solution :
Here each person (pigeon) is assigned the day of the week
(pigeonhole) on which he and she was born since there are eight
people and only seven days of the week, the pigeonhole principle.
Tells us that at least two people must be assigned to the same day.
Example 2:
Consider the area shown it is bounded by a regular hexagon.
Whose sides have length 1units. Show that if any seven points are
chosen with in this area then two of them must be on further apart
then 1 unit.
Solution:
Suppose that the area is divided in to six equilateral triangles.
As shown in figure 1.1
1
6

If seven points are chosen we can assign each one to a

triangle that contains it.
If the point belongs to several triangles, assigns it arbitrarily
to one of them. The seven points one assigned to six triangles so by

116

pigeonhole principle, at least two points must belong to the same

triangle. These two can not be more then 1 unit apart.
Example 3 :
Five points are located inside a square whose sides are of
length 2. Show that two of the points are within a distance
each other.

2 of

Solution :
Divide up the square into four square regions of area 1 unit.
as indicated in figure 1.2.
1
1
1

fig.1.2
By Pigeonhole principle, it follows that at least one of these
regions will contain at least two points. The result now follows since
two points in a square of radius 1.can not be further apart then length
of the diagonal of the square is which (by Pythagoras theorem) 2 .
Example 4 :
Show that if any five numbers from 1 to 8 are chosen, then
two of then will add to 9.
Solution :
Constructs four different sets each contains two numbers that
add to 9, as follows A1 1,8 , A2 2, 7 , A3 3, 6 , A4 4,5 each of
the five numbers chosen will be assigned to the set that contains it
.Since there are only four sets. The pigeonhole principle tells that
two of the chosen numbers will be assigned to the same set. These
two numbers will add to 9.
Example 5 :
Fifteen children together gathered 100 nuts. Prove that some
pair of children gathered the same numbers of nuts.

117

Solution :
Now to prove that we use method of contradiction.
Suppose all the children gathered a different numbers of
Then
the
fewest
total
number
is
0 1 2 3 4 5 6 14 105 , but this is more then 100.
Which is contradiction to our assumption. There fore at least pair of
children gathered same number of nuts.
nuts.

Example 6 :
Show that in any set of 10 integers there are at least pair of
integers who have same remainder when divided by 9.
Solution :
Set of 10 integers, when it divide by 9, lie in the same residue
classes of modulo 9. i.e. the remainder is 0,1,2,3,4,5,6,7,8. Here
there will be 9 remainder and 10 integers. There fore by pigeonhole
principle, at least one integer has same remainder.
Example 7 :
Any 7 numbers are chosen from 1-12. Show that,
(i). Two of them will add to 13.
(ii). There are two respectively prime integers.
(iii). There are two integers such that 1 is a multiple of the other.
Solution :
(i). We form the box different sets (boxes) each containing two
integers from 1-12 with their sum as 13 as follows.
{1, 12}
{4, 9}
{2, 11}
{5, 8}
{3, 10}
{6, 7}
These six sets are the boxes and the 7 integers chosen from 112 are objects each of these 7 integers will be arranged if the set that
contains it since there are only 6 boxes and 7 objects By Pigonhole
principle, two of the selected numbers will be arranged to the same
box, hence there sum will be 13.
(ii). Two numbers are said to be relatively prime if their G.C.D is 1.
We form 6 different boxes containing two integers form 1-12
such that they are respectively prime as follows,
{1, 2}
{7, 8}
{3, 4}
{9, 10}
{5, 6}
{11, 12}
Since each set contains consecutive integers.

118

They are relatively prime each of these 7 integers will be

assigned to the set that contain it these 6 sets are boxes and 7
integers are chosen from 1-12 are objects.
Since there are only 6 boxes and 7 objects 6 < 7.
By pigeonhole principle selected numbers will be arranged to
the same box that is their exists two numbers which are relatively
prime.
(iii). Let A1, A2,.A7 be seven chosen numbers from 1-12.
We known that any integers n can be written in the form n =
m where k 0 km. is an odd number.

120 1
221 1
320 3
422 1
520 5
621 3
720 7
a1 = 2k1 x1 , a2 = 2k2 x2. a7 = 2k7 x7
Where ki 0.
Xi is an odd integer..i=1,2,.,7.
Each ai is assigned to the odd integer xi where xi is such that ai =
2kl.ai.
Where ki 0.
Corresponding to u chosen integer a1, a2,.,a7 we have 7 odd
integers.
x1, x2,.,x7 but we have only 6 odd integers from 1-12.
By Pigonhole principle ais corresponding to the same xi.
i.e.7i ,j such that I j
1 i 7
1 j 7 and xi = xj
ai = 2ki.xi
aj = 2kj.xj
aj = 2kj.xi [xi = xj]

119
kj
if xi= xj then 2

kj xi
2

2ki

2kixi

a
j

ai
ai is a multiple of aj.
kj
if xi xj then 2

kj xi
2

a
j

2ki

2kixi

ai

aj is a multiple of ai.
There for their exist two integers such that one is a multiple of
another.
Example 8.Show that in any set of 12 integers there are 2 whose
difference is divisible by 11.
Solution :Let A1, A2,.A12 be 12 selected integers by division
algorithm, 7 unique qi and ri.
s.t. Ai = qi + ri where 0 ri 11
i.e. 0 ri 10
each qi is assigned to its remainder ri.
A1, A2,.A12 corresponds to r1, r2,.r12 but use
have only 11
Remainders {0,1,2,,10}
When a no. is divided by 11
The remainders are 12 i.e. r1,r2,.r12 but possible
remainders are 11.
i.e. 0,1,..11
By Pigonhole principle, two remainders are same i.e.i , j
s.t.i

j, ri = rj

1 i 12,
1 j 12

ai = 11qi + ri
aj = 11qj + rj
=11qj + ri (ri = rj)
i.e.

120

two whose difference is divisible by 11.

Two whose difference is a multiple of 11.

7.3 THE EXTENDED PIGEONHOLE PRINCIPLE

If there n -pigeons are assigned to m -pigeonholes, then one of the
n 1
pigeonhole must contain at least
1 pigeons.
m

n 1
Proof: If each contain number more then
pigeons, then there
m
n 1
n 1 n 1.
are at most
m
m
m
A pigeon in all this contradicts our assumption. So one of the
n 1
pigeonholes must contain at least
1 pigeons.
m
Example 9: Show that if 30 dictionaries in a library contains a total of
61,327 pages, then one of the dictionaries must have at least 2045 pages.
Solution: Let the pages be the pigeons and the dictionaries are the
pigeonholes. Assigns each to the dictionaries in which it appears then by
the extended pigeonhole principle are dictionary must contain at least
61,327 1
61,326
1 2045 pages.

1
30
30

Example 10: Show that if any 29 people are selected then one may choose
subset of 5. So that all 5 were born on the same day of the week.
Solution: Assign each person to the day of week on which he and she was
born. Then n 29 persons are being assigned to m 7 pigeonholes. By
the
extended
pigeonholes
principle
at
least
n 1
29 1
28
1 5 persons.

1
1
7
m
7
There fore 5 persons must have been born on the same day of the week.

Check Your Progress :

1. Show that if there are seven numbers from 1 to 12 are chosen
then two of them will add to 13.

121

2. Let T be an equilateral triangle whose sides has length 1 unit.

Show that if any five point are chosen lying on inside T. Then
two of them will be more then

1
unit apart.
2

3. Show that if any Eight positive integer are chosen two of them
will have the same remainder when divided by 7.
4. Show that if seven colors are used to paint 50 bicycles at least
eight bicycles must have the same colors.
5. All 82 entering student of a certain high school take courses in
English, History, Maths and science. If three section of each of
these four subjects. Show that there are two students that have all
four classes together.
6. Nineteen points are chosen inside a regular hexagon whose side
length 1. Prove that two of these points may be chosen whose
distance them is less then

1
.
3

7. In any group of 15 people there are at least three born on the

same day of the week?

7.4 LET US SUM UP

In this chapter we have covered the following points:
If m n then there cannot be an injection from n to m .
Pigeon hole Principle.
Extended Pigeonhole Principle and Application of pigeon hole
principle.

7.5 UNIT END EXERCISES

1. Prove that, if m n then there cannot be an injection from n
to m .
2. Write the statement of pigeon hole principle and explain with
example.
3. 10 people want to go to the movies, and there are only 7 cars,
then at least more then one person in the same car.
4. Prove that among the 51 positive integers less then 100. There is
a pair whose sum is 100.

122

5. There are 33 students in the class and sum of their ages 430 year.
Is it true that one can find 20 students in the class such that sum
of their ages greater 260?
6. Show that in any set X of people, there are two members of X
who have the same number of friends in X.
7. Seven darts are thrown onto a circular dartboard of radius
10units. Can we show that there will always be two darts which
are at most 10 units apart?
8. Nineteen darts are thrown onto a dartboard which is shaped
as a regular hexagon with side length of 1 unit. Can we prove
that there are two darts within

3
units of each other?
3

9. How many friends must you have to guarantee that at least

five of them will have birthdays in the same month?
10. Show that there must be at least 90 ways to choose six numbers
from 1 to 15 so that all the choices have the same sum.

123

8
GRAPH THEORY-I
Unit Structure :
8.0 Objectives
8.1 Introduction
8.2 Application of graphs
8.3 Basic definitions and types of graphs
8.4 Subgraphs and Isomorphisms
8.5 Operations on graph
8.6 Let us sum up
8.7 References

8.0 OBJECTIVES

Fundamental concepts of graphs

Types of graphs
Isomorphism of graphs
Concept of connectedness in graph

8.1 INTRODUCTION
Graph theory is a subject where no previous knowledge is
assumed. In this subject the focus is on understanding the structure
of graphs and the techniques used to analyse problems in Graph
theory. This subject have many applications in different areas right
from computing to social sciences and to natural sciences. One of
the standard ways of maintaining a graph in the memory of computer
is by means of its adjacency matrix. In simple words a graph is
collection of points and a collection of pairs of points (edges). Some
of the graphs look like.

Fig. 8.1

124

8.2 APPLICATION OF GRAPHS

A graph can be used to represent almost any physical
situation involving discrete objects and relationship among them.
Some of examples given below are from among hundreds of
application.
1. Konigsberg Bridge Problem :This exciting problem is said to have given birth to graph
theory. The city of Konigsberg is located on the Pregel river in
Prussia. The city occupied the island of Kneiphopf (A) plus areas on
both banks. These regions were linked by seven bridges as shown
below in Fig 8.2 (a). The citizens of Konigsberg had a problem to
start from their home, cross every bridge exactly once and return
home. This problem was represented using dots for land masses and
curves for bridges Fig. 8.2 (b).
C
Pregel

river

D
Fig. 8.2 (a)
The bridges of Konigsberg
C

Fig. 8.2 (b)

Graphical representation of Konigsberg bridge problem
Euler first represented this situation by means of graph and proved
that the solution for this problem does not exist using Eulerian
graphs.

125

2. Travelling salesman problem :This problem is stated as follows : A salesman is required to

visit a number of cities during a trip. Given the distance between the
cities, in what order should he travel so that he travels as minimum
as possible?
In the graph theory, the cities are represented by vertices and
the roads by edges. There is a real number associated with each
edge ei in a graph. Such a graph is called weighted graph. w(ei ) is
weight of edge ei .
This problem amounts to finding minimum weight
Hamiltonian circuit in a weighted Hamiltonians graph. No efficient
algorithm for solving this problem is known. However, it is possible
to obtain reasonably good but not necessarily optimal solution.
3. Four colour problem :The four colour problem is a partitioning problem.
Partitioning is applicable in many practical problems such as coding
theory, partitioning of logic in digital computers and state reduction
of sequential machines.
This problem is related to colouring of a map. A map is a
partition of the plane into connected region. Can we colour the
regions of given map using atmost four colours so that neighbouring
regions have different colours?
In graph theory, the vertex is considered as a region and an
edge represents regions sharing a boundary. The problem is whether
the resulting graph have chromatic number atmost 4. A graph can be
drawn in a plane without crossing edges, such graphs are planar
graphs.
The four colour problem was posed in 1852 and was settled
by Appel and Haken in 1972. A computer free proof of this problem
is still to be found.

126

8.3 BASIC DEFINITIONS AND TYPES OF GRAPH :

The Fig. 8.3. Shows some of the graphs.
v1

v1

e2
e1

e1
v2

v2

e2

Fig. 8.3.1 (a)

v1

v3

v3

e3

v5
v3
v4

e4

Fig. 8.3.1 (b)

v2

Fig. 8.3.1 (c)

In Fig. 8.3.1 (a) there are points v1,v 2 ,v3 and the line
segments e1, e2 . Line segment e1 joins points v1 and v 2 while line
segment e2 joins points v 2 and v3 .
In Fig. 8.3.1 (b) point v 2 is joined to itself by a loop e4 .
There are multiple line segments between points v1 and v 3 . In Fig.
8.3.1 (c). There are no line segments but the points v1, v 2 , v3 , v 4 , v5 .
All the above figures are the examples of graphs. Each graph
consists of certain number of points called vertices and some pairs of
points joined by line segments which are called as edges.
8.3.1 Definition : Graph, order and size of graph
A graph G is a pair (V, E) where V is nonempty finite set of
vertices and E is family of unordered pairs of elements of V called
edges. V is vertex set of G, E is edge set of G. The number of
elements in V is called order of G. It is denoted by |V|. The number
of edges in graph G is called size of G and denoted by |E|.
In Fig.8.3.1 (a),

V=v1, v 2 , v3 , E = e1, e2 , |V| = 3 & |E| = 2.

In Fig. 8.3.1 (b), V = v1, v 2 , v3 , E= e1, e2 , e3 , e4 , |V| = 3 and

|E| = 4
In Fig. 8.3.1 (c), V= v1, v 2 , v3 , v 4 , v5 , E = 0 (empty), |V| = 5
and |E| = 0

127

8.3.2 Definition : Simple graph, Multigraph, Directed graph

A simple graph G (V, E) consists of V, a nonempty set of
vertices and E, a set of unordered pairs of distinct elements of V
called edges. Fig. 8.3.2 (a).
A multigraph G(V, E) consists of set V of vertices, a set E of
edges including multiple edges and loops. Fig. 8.3.2 (b).
A directed graph G(V, E) consists of V, nonempty set of
vertices and E which is family of ordered pairs of elements of V
which are directed edges. Fig. 8.3.2 (c).

Fig. 8.3.2 (a)

Simple Graph

Fig. 8.3.2 (b)

Multiple Graph

Fig. 8.3.2 (c)

Directed Graph

8.3.3 Definition : Incidence, Adjacent, Degree, Pendant vertex

See the Fig. 8.3.1 (a) the edge e1 joins vertices v1 and v 2 .
The vertices v1 and v 2 are called end vertices of edge e1 . The edge
e1 is said to be incident on vertex v1 and vertex v 2 . The vertices
v1 and v 2 are adjacent vertices. Thus, two vertices are said to be
adjacent if they are the end vertices of the same edge. Similarly, two
nonparallel edges are said to be adjacent if they are incident on a
common vertex. In Fig. 8.3.1 (a), e1 and e2 are adjacent edges.
The number of edges incident on vertex v, with self loop counted
twice is called degree, d(v), of vertex v. In Fig. 8.3.1(a)
d v 2 = 2 . In Fig. 8.3.1(b) d v 2 = 3 .
The vertex having no incident edge is called an isolated vertex.
Isolated vertices are vertices with zero degree. A graph with all
isolated vertices is called null graph. Fig.8.3.1 (c).
The vertex of degree one is called pendant vertex.
8.3.1(a), vertices v1 and v 3 are pendant vertices.

In Fig.

128

Theorem 8.3.1 :
The Handshaking Theorem
The sum of degrees of all vertices in graph G is twice the number of
edges in G.
Proof :Let G be a graph with e edges and n vertices v1, v 2 ,..., v n . To show
n

that

d vi = 2 e

(3.1)

i=1

Each edge contributes two degrees to the sum of degree of

vertices as each edge is incident with exactly two vertices. This
means that the sum of degrees of the vertices is twice the number of
edges. A loop at a vertex also counted twice.
n

Thus,

d vi = 2 e
i=1

For example consider the graph in fig. 8.3.3

v1

v2

v3

v4

Fig. 8.3.3
d v1 + d v 2 + d v3 + d v 4

= 2 + 2 + 3 + 1 = 8 = 2 4 = twice the number of edges.

Theorem 8.3.2 :
The number of vertices of odd degree in a graph is always even.
Proof :Consider the vertices of odd degrees and even degrees separately.
n

The sum of degrees of all vertices is even i.e.

d vi = 2 e , in

i=1

graph G with n vertices v1, v 2 ,..., v n and number of edges e.

n

Also, d vi =
i=1

d vi + d v k
odd

even

(3.2)

129

The left side of equation 3.2 is expressed as sum of two sums,

each sum taken over vertices of odd and even degrees respectively.
n

d vi = 2 e

Since
side i.e.

i=1

d v k = 2s , both are even,

(say)

even

we get

and the second expression on right hand

d vi = 2 e d v k
odd

even

= 2e 2s
= 2 (e s)

i.e.

d vi = an even number

(3.3)

odd

Thus, because of equation 3.3, each d vi is odd.

The total number of terms in the sum must be even, to make
the sum an even number. Hence the theorem.
Types of Graphs
(a) Complete Graphs :
A simple graph in which every pair of distinct vertices is
adjacent is called a complete graph. Denote complete graph on nvertices by k n .
Note :
1) In K n , every vertex is adjacent to all the remaining (n 1)
vertices.
2) The degree of each vertex of K n is n 1.
Some examples of complete graphs.

K1

K2

K3

K4

Fig. 8.3.4.1 complete graphs

K5

130

(b) Regular Graphs :

If all the vertices in a graph G are of the same degree, then G
is called regular graph. If degree of each vertex is n then G is said
to be regular of degree n.
Remark :
(1) All the complete graphs are regular.
(2) A regular graph of degree 3 is known as cubic graph.
The graph of Fig.8.3.4.2 is Petersens graph. In this graph
there are ten vertices and degree of each vertex is 3. It is also an
example of cubic graph.
v1

v5

v10

v
6

v2
v
7

v
9

v8
v3

v4

Fig. 8.3.4.2 Petersens Graph

(c)
Bipartite Graph: If the vertex set V of a graph G can be
partitioned into two disjoint subjects say V1 and V2 such that any
edge in G joins a vertex of V1 to a vertex of V2 , then G is called
bipartite graph.

a
f

d
(a)

(b)

Here V1 = a, c, e

V2 = b, d, f

Here V1 = a, c, e

V2 = b, d, f

Fig. 8.3.4.3 Bipartite Graphs

131

Remark :
(1) In bipartite graph, it is not necessary that every vertex in V1 is
adjacent with every vertex in V2 . But if it is so and G is simple
graph then G is called complete bipartite graph. A complete
bipartite graph is denoted by Km,n where V1 = m and V2 = n .
(2) Km,n has mn edges as each vertex in v1 joins every vertex in
V2 . The Fig. 7.3.4.3 shows K 3,3 graph.
(3) The graph K1,n is known as star. For example, Fig. 7.3.4.4 is
Star Graph K1,7 .

Fig. 8.3.4.4
Star Graph; K1,7
Check your progress :
1. Draw all simple graphs on one, two, three and four vertices.
2. Describe the graph G in the diagram i.e.
i) Find vertex set V(G) and |V(G)|
ii) Find edge set E(G) and |E(G)|
iii) Find the degree of each vertex.
iv) Verify Handshaking theorem for given graphs.

(a)

(b)

3. Draw a simple graph, multiple graph and directed graph on seven

vertices.
4. Show that number of odd degree vertices in a graph is always
even.

132

5. Consider the graph in following diagram.

v
1
v4

e
4

v2

e
1
e
2

v3

e
3

v9
e
5

v
8

e
7
v5

e6

v
7

v
6

Find (i)
end vertices of edges e5 ,e4 ,e3
ii) edge incident on vertices v1 and v 2 , v 4 and v5 , v 2 and v3
iii) pair of adjacent vertices (any three)
iv) pair of adjacent edges (any three)
v) isolated vertices
vi) pendant vertices
6. Draw following graphs
i) K 4
ii) K5
iii) Regular graph of degree three
iv) Petersens graph
v) K 2,3
vi) K5,3
7. Justify whether the following statements are True or False.
i) A circle with radius one and centre at origin is a graph.
ii) In a simple graph on p vertices the degree of each vertex is
atmost p 1 .
iii) There does not exist a graph on give vertices whose degrees are
4, 1, 2, 2, 3.
iv) The number of vertices of odd degree is odd.
v) There does not exist a complete graph on n (n > 2) vertices which
is bipartite graph.

8.4 SUBGRAPHS AND ISOMORPHISM

A) Subgraphs
8.4.1 Definition :
A graph H is called a subgraph of a graph G if all the vertices
and the edges of H are also vertices and edges of G. In other words,
V(H) V(G) and E(G) E(H) .

133

For example,
Consider Petersens graph P given in Fig. 8.3.4.2. The following
graphs P1, P2 , P3 are subgraphs of P.
v1

v10

v
6

v5

v10

v
7
v
9

v8

v2

v
6
v
7

v
9

v8
v3

v4

P1

v2

v5

v8

P2

P3

Fig. 8.4.1
Subgraphs of Petersens Graph
The graph P0 is not a subgraph of P as shown in Fig. 8.4.2
because the edge set of P0 is not contained in E(P).
v1

v
7

v4
v3

Po

Fig. 8.4.2
Graph which not subgraph of Petersens Graph
Note : Subgraph of bipartite graph is bipartite, because all the edges
in the subgraph are edges of original graph.
8.4.2 Definition :
A subgraph H of a graph G is called spanning subgraph of G if H
contains all the vertices of G. In other words V G = V H .
Let G be a graph as shown in Fig. 8.4.3, then graphs H1, H 2 , H3 of
Fig. 8.4.4 are spanning subgraphs of G.

134
v5

v1
v2

v3

v4

Fig. 8.4.3
v
1

v5

v1
v2

v4

v3

H1

v1

v5
v2

v3

v5

v2
v
4

H2

v3

v4

H3

Fig. 8.4.4
Spanning subgraphs
8.4.3 Definition :
Let G be graph with vertex set V. Let any set S be subset of V, the
induced subgraph <S> is defined as the maximal subgraph of G with
vertex set S.
Note : There is no subgraph of G with vertex set S that contains <S>
properly.
Example :
1) Consider Petersens graph P in Fig.8.3.4.2 and its subgraphs
shown in fig. 8.4.1. In this case P1 is subgraph of P induced by
S = v6 , v7 , v8 , v9 , v10 i.e. P1 =< S > . Note that P2 is not an
induced subgraph of G.
2) Consider Fig. 8.4.5 (a). If we take S = v1, v 2 , v 4 then <S> = H1
is not induced subgraph because the graph has vertex set S but
H1 H 2 i.e. H1 is not a maximal subgraph of G with vertex set S.
Thus, H 2 is induced subgraph of G.

135
v1

v1
v4

v2

v1

v4

v2

v3

(a) H1

v4

v2

(b) H2

(c)

Fig. 8.4.5
Induced subgraph
B) Isomorphisms of Graphs
8.4.4 Definition :
Two graphs G1 and G 2 are said to be isomorphic if there
exists a one-one and onto map T: V G1 V G 2 such that if
v1, v 2 V G1 , then the number of edges joining v1 and v 2 is also
the same as the number of edges joining T v1 and T v 2 .

Note that isomorphism of two graphs preserves adjacency and nonadjacency of any two vertices.
Example :
Consider following two graphs G1 and G 2 in Fig. 7.4.6.
v3

v1

v2

v5

v4

v4

v5

v3

v2

G1

v1

G2

Fig. 8.4.6
Isomorphic graphs
Define a map from V G1 to V G 2 as follows : v1 u1 , v 2 u 2 ,
v3 u 3 , v 4 u 4 , v5 u 5 . This map is an isomorphism. G1 and
G 2 are isomorphic graphs.

136

Remarks :
Two isomorphic graphs have the same number of vertices,
same number of edges and also same number of vertices of given
degree. However, these conditions are only necessary not sufficient.
For example, consider graphs G1 and G 2 given in fig. 7.4.7. They
have same number of vertices, same number of edges and also each
of them have exactly one vertex of degree three and four, two
vertices of degree two and five vertices of degree one. Yet, they are
not isomorphic because under the isomorphism, v 2 and v5 are not
adjacent in G1 where as in G 2 , u 4 and u5 are adjacent.
d(v 2 ) = 3 and d(u 4 ) = 3

d(v5 ) = 3 and d(u 5 ) = 3 Thus, v5 u4 and v5 u 5

v8

v1

v2

v
6
v3

:G
1

v5

v4

v
7

v
9

u3

u
4
u

6
:G
2

Fir. 8.4.7 Non-isomorphic Graphs

C) Matrix representation of Graphs
Graphs can also be represented by matrices. This method is
most suitable for computer processing. The two representations of
graphs using matrix is by incidence matrix and adjacency matrix.
8.4.5 Definition : Incidence matrix
Consider graph G without loops. Let V = v1, v 2 ,...v m and

E = e1, e2 ,..., en Define m x n matrix I(G) as follows :

I(G) = aijmn

where aij = 1, if e j is incident at vi

= 0, otherwise

137

Example of incidence matrix :

Consider graphs and its incidence matrix, in Fig. 8.4.8
e

v1

e1 e2

v4
e

e
4

v
2

5
v3

v1 1
v 2 1
I(G) v3 0

v4 0

v3
e

v1

e
e

0
1
1
0

e3 e 4
1
0
0
1

0
1
0
1

e5
1
0
1
0

e1 e2 e3 e4
e
3

v2

v1 1 1 0 0
I(H) = v 2 0111
v3 1 0 1 1

Fig. 8.4.8 Incidence matrix

Note :
1) Incidence matrix is defined for graphs without loops.
2) All entries in incidence matrix is zero or one. Such matrix is
called binary matrix.
3) There are exactly two ones in each column.
4) The sum of the entries in any row is the degree of the
corresponding vertex.
5) A row corresponding to a vertex of degree zero contains all
zeros.
6) Two graphs G1 and G 2 are isomorphic if and only if their
incidence matrices I G1 and I G 2 differ only by permutation
of its rows and columns.
8.4.6 Definition : Adjacency matrix
Let G be a graph with n vertices, say v1, v 2 ,...v n and not
having multiple edges. Then the matrix A(G) = x ij of order n n,

138

called the adjacency matrix of G is defined as the matrix whose ijth

entry is given by,

x ij = 1 , if there is an edge joining ith and jth vertices.

= 0, otherwise
Example of Adjacency matrix:
Consider graphs and their adjacency matrix, in Fig 8.4.9
v
1

v
5

v1 v 2
v1 0 0
v2 0 0

A(G ) v3 0 1

v4 0 1
v5 0 0

v2

v3
v4

v3

v4

0
1

0
1

1
1
1

1
0
1

v5
0
0
1

1
0

a b c d
a 0
b 1
A(H) =
c 0

d 0

1 0 0
0 1 1
1 0 1

1 1 0

H
Fig. 8.4.9 Adjacency matrix
Note :
1) Adjancency matrix is a symmetric square matrix with entries
either zero or one.
2) If G has no loops then the sum of the entries along any row or
any column is the degree of the corresponding vertex.
3) If G1 and G 2 are graphs without multiple edges then G1 is
isomorphic to G 2 if and only if A(G1 ) can be obtained from
A(G 2 ) by applying permutation of rows and of corresponding
columns.

139

Check your progress

1. Draw a bipartite graph whose subgraph is bipartite.
2. Draw any two spanning subgraphs of following graphs.
v
1

v2

v3
v3

v2
v
1
v
6

v5

v
4

v5

v
4
v5

(a)

v
6

(b)

3. Draw any two induced subgraphs of the graphs given in

Exercise 2.
4. Give any two examples of non-isomorphic graphs.
5. Prove that following graphs are not isomorphic.

6. Determine whether given pair of graphs are isomorphic. If not,

then justify. If isomorphic then show mapping.

a)

b)

140

c)

7. Show that there are four non-isomorphic simple graphs on three

vertices.
8. Draw graphs corresponding to following incidence matrix.
e1 e 2 e3 e 4
v1 1
v 2 0
v 3 1

i)

1
1
0

0
1
1

a b c d
v1 0
v2 0

v3 0

v4 1
v5 0

v6 1

ii)

0
1
1
e f

0 0 1 0 1
0 0 0 1 1
0 0 0 0 0
1 1 0 1 0
0 1 1 0 0
1 0 0 0 0

g h
0 0
1 1
0 1

0 0
1 0

0 0

9. Write incidence matrix for simple graphs

v
1

i)

v
e

e1

v5

e
v3

e
b

v
4

v2

ii)

e4

e
2
v2

v
1

f
v3
g

h
c

10. Draw Graph with given adjacency matrix.

i)

0 1 0
1 0 1

0 1 0

ii)

0
0

0 1 1
0 1 0
1 0 1

1 1 0

v5
d
v
4

141

11. Represent given graphs using adjacency matrix.

i)

ii)

12. Justify whether following statements are true or false.

i) There are eleven non-isomorphic simple graphs on four vertices.
ii) Let H be subgraph of G. If u.v are non adjacent vertices in H then
they are also non-adjacent in G.
iii) If H is induced subgraph of G and if two vertices are not adjacent
in H then they are also not adjacent in G.
iv) Subgraph of subgraph of G is a subgraph of G.
v) An induced spanning subgraph of G is G itself.
vi) Adjacency matrix is symmetric square matrix.
vii)

Incidence matrix is defined for graphs with loops.

viii) A row corresponding to a vertex of degree zero in incidence

matrix contains all non zero entries.

8.5 OPERATIONS ON GRAPHS

8.5.1 Removal of vertex or edge
Let G be a graph with V = v1, v 2 ,..., v n . The removal of
vertex vi results in a subgraph G of G consisting of all vertices of G
except vi and all edges of G which are not incident with vi .
G = G vi is maximal subgraph of G with vertex set,

S = v1, v 2 ,..., vi1, vi+1,..., v n

Example :
Consider the graphs G v 4 and G v3 obtained by removing vertex
v 4 and vertex v3 respectively in Fig. 7.5.1.

142
v
1
v
1

e
3

e
1

v
5

e
2

v2

v2

v2

e4

v
6

e
5
e6

v
e 3
7

v3

v4

v
5

v
5

G - v4

v4

G - v3

FIG. 8.5.1(a) Removal of vertex

If E(G) = e1, e2 ,..., ek then removal of an edge say ei from
G results in graph G where V(G) = V(G ) and E(G ) contains all
edges of G except ei .
Example :
G e3 and G e4 are two graphs obtained by deleting edge e3 and
edge e5 respectively from graph G given in Fig. 8.5.1.
v
1
e
1

v2
e
2

e
4

e5

e
1

v
5

e
7

v3

e6

v4

e
2

v3
v
5

G- e3

e
3

v
1

v
6

e5
e6

e
7

v2
v
6

v4

G - e4

Fig. 8.5.1(b) Removal of edge

Note : When a vertex is removed or deleted all edges incident on
that vertex is removed but when a edge is removed we do not
remove vertices incident at that edge.
8.5.2 Complement of a graph
Let G = V(G), E(G) , then complement of G is graph G with
vertex set V(G) where in two vertices are adjacent if and only if they
are not adjacent in G.

143

Example :
See Fig. 8.5.2(a) for graph G and its complement.
v1
v2

v1

v2
v
5

v
5

v3
v3

v4

v4

G : Complement of G

Fig. 8.5.2(a) Complement of a graphs

A graph G is self complementary if it is isomorphic to its
complement G as shown in Fig. 8.5.2(b).
v1

v1

v
5

v2

v
4

v
3
G

v
5

v2
v
4

v
3
G

Fig. 8.5.2(b) Self complementary Graphs

G and G in Fig. 8.5.2(b) are isomorphic under the map
v1 v1 , v 2 v3 , v5 v 4 , v3 v5 , v 4 v 2 .
8.5.3 Union of two graphs
Let G1 V1,E1 and G 2 V2 ,E 2 be two graphs. The union
G1 G 2 of G1 and G2 is defined as graph where vertex set is
V1 V2 and edge set is E1 E 2 .

144

Example :
Union of two graphs i.e. illustrated in Fig. 7.5.3.
e
4

v1
e1
v3

v2

f1

v4

e
3

f2

u
1

e
3
e2

u2

v3

G1

f3

v1

v4

G2
v1

v2

e1
f3
v3
e2

e
3

u1
f1

v4
f2

G1G 2

Fig. 8.5.3 Union of two graphs

8.5.4 Intersection of two graph
Let G1 V1,E1 and G 2 V2 ,E 2 be two graphs then the
intersection G1 G 2 of G1 and G 2 is graph consisting of only those
vertices and edges which are both in G1 and G 2 .
Example :
Illustration of G1 G 2 is G 2 as shown in Fig. 7.5.3.

v1

e
3
v4
v
3
G1G 2

Fig. 8.5.4 Intersection of graph

145

8.5.5 Ring sum of two graphs

The ring sum of graphs G1 V1, E1 and G 2 V2 , E 2 is
consisting of vertex set V1 V2 and edges that are either in G1 or in
G 2 but not in both and is denoted by G1 G 2 .
Example :
Taking G1 and G 2 from Fig. 8.5.3, the ring sum G1 G 2 is
e
4

v
1
e
1

v2

u1
f
1

f3

v3

e
2

f2
u

v4

Following is one more example which shows G1 G 2 , G1 G 2 and

G1 G 2 . Fig. 8.5.5(a)
v

e
2

e
1
v2

v
1

v3

e
3

e
4

e5

f
1

e
1

e3

v4

v5

e6

G1

e
1

f4
f3

G2

v
1
e
1

e
2

1
v2

v3

G1G 2

v
5

v3

v2
e3

v
6

f2

v2

v4

e3

e6

G1G 2

f
1
f2
v3

f3
f4

v
6

146
v

f
1
e
2

f2

v2

v3

e
4

e5

v4

e6

f3

v
6

f4

v5

G1G 2

Fig. 8.5.5(a) Intersection, Union and ringsum of two graphs

8.5.6 Fusion and Contraction
A pair of vertices v1 and v 2 in graph G is said to be fused if
the two vertices are replaced by single vertex v such that every edge
that was adjacent to either v1 or v 2 or both is adjacent to the new
vertex v . The fusion does not alter the number of edges in the
graph but number of vertices is reduced by one.
Example :
Fig. 8.5.6.1(b) is graph obtained by fusion of vertices v1 and v 2 in
Fig. 8.5.6.1(a).
v
1

v = ( v1, v2 )

v2

v
6

v5

v5

v
4

v3

v
6

(a)

v
4

v3

(b)
Fig. 8.5.6.1 Fusion of vertices

Contraction of an edge in a graph is obtained by taking an edge with

end vertices u and v, and contracting it. In other words, removing e
and identifying u and v in such a way that the resulting vertex is
incident to those edges (other than e) which were originally incident
to u or v.

147

Example :
Fig. 8.5.6.2(b) is a graph obtained by contraction of an edge e in
graph of Fig. 8.5.6.2(a).

(u,v)

v1

v4
v2

v4

v1
v3

v2

v3

(a)

(b)

Fig. 8.5.6.2 Contraction of an edge

8.5.7 Sum of two graphs
Let G1 V1, E1 and G 2 V2 , E 2 be two graphs where V1 V2
then sum of G1 and G 2 denoted by G1+ G 2 is defined as graph
whose vertex set is V1 V2 and consisting of all edges, which are in
G1 and G 2 , and the edges obtained by joining each vertex of G1 to
each vertex of G 2 .
Example :
Fig. 8.5.7 show the sum G1+ G 2 .
v
1

v
1

v2

v4

u
v4

v3

v3

G1

G2

Fig. 8.5.7 Sum of two graphs

G1 + G 2

v2

148

8.5.8 Line graph

The line graph L(G) is a simple graph G whose vertices are in
one-one correspondence with the edge of G, two vertices of L(G)
being adjacent if and only if corresponding edges of G are adjacent.
Example :
See Fig. 8.5.8.
v

e1
v2

e
2

e1

v5

e4

5
e

v
4

v3

e4

L(G)
Fig. 8.5.8 Line Graph

8.5.9 Cartesian product of G1 and G2

Let G1 V1, E1 and G 2 V2 , E 2 . Then carterian product of
G1 and G 2 . G1 G 2 is graph with vertex set V specified by putting
u adjacent to v if and only if
i)
u1 = v1 and u 2 v 2 E 2 or
ii)

u 2 = v 2 and u1v1 E1

Example :
See Fig. 8.5.9
u

v
1

u
2

G1
( u ,u )
1 2

( v ,u )
1 2

v
2

w2

G2
( u ,v )
1 2

( v ,v )
1 2

( u ,w )
1 2

( v ,w )
1 2

G1 xG 2

Fig. 8.5.8 Cartesian product of graphs

149

Remark :
If V1 = u1, v1 and V2 = u 2 , v 2 , w 2 then cartesan product of sets
V1 V2 is
V1 V2 = u1, u 2 , u1, v 2 , u1 , w 2 , v1, u 2 , v1 , v 2 , v1 , w 2

The

points in V1 V2 are vertices of G1 G 2 .

At this juncture, let us take a break and do some problems and
check if we have understand the basic concepts of graphs.
Check your progress
1.

Consider the graphs given below

v
1

e1

v2

e3

v
1

v3

v
4

G1

e
6
e
5

e
2

v2

e1

e
7

v5

e
4

v3

e
3
v
4

G2

Draw the graph

i)
G1 v 2
ii)
G1 v1
iii)
G 2 v5
iv)
G 2 v3
v)
G1 e4
vi)
G1 e3
vii)
G 2 e3
viii) G 2 e7
2.

Draw a graph which is self complementary.

3.

Draw complement of graphs given in exercise 7.5.1.

4.

Draw the graphs G1 + G 2 if G1 and G 2 are given as follows :

150
v
1
v
1

e1
e

v2

v3

e
e

e1

f3

f6

v3

v5

e
3

f1

v2

u
1

f2
f5

f4
u

v
4

u
3

G1

G2

5.

Draw the graph obtained by fusion of vertices v1 and v 2 in

graphs given in exercise 7.5.1.

6.

Draw the graph obtained by contracting the edge e1 in the

graphs given in exercise 7.5.1.

7.

Let G1 V1,E1 and G 2 V2 ,E 2

be two graphs with

V1 = u1, v1 , E1 = u1, v1 while V2 = u 2 , v 2 , E 2 = u 2 , v 2 .

Draw the graph of Cartesian product of G1 and G 2 .

8.

Justify whether following statements are True or False.

i)

The ring sum of two complete graphs is complete.

ii) If

G,

H,

are

G ( H K ) = (G H ) (G K )

simple

graphs

then

iii) The complement of every graph and its line graph are
isomorphic.
iv) Adjacency matrix of simple graph and its complement is
different.
v)

G1 and G 2 is regular if G1 V1,E1 , G 2 V2 ,E 2 are two

graphs with V1 = 3 , E1 = 2 , V2 = 4 and E 2 = 3 .

8.6 LET US SUM UP

In this chapter, we have learnt basic concepts of graph theory
to start with. The understanding of graph theory starts with this basic
concepts as vertex, edge, edge set and vertex set. The different
operations on graph lead to a big family of graph. Isomorphism of
graphs tells vs about graphs which are isomorphic.

151

8.7 REFERENCES
1. Discrete Mathematical structures by Kolamn, Busby and Ross.
Pearson education.
2. Introduction to graph theory by Douglas B. West.
3. Discrete Mathematics and its applications by Kenneth. H.Rosen.
McGraw Hill edition.
4. Graph theory by Frank Harary. Narosa Publication.
5. Discrete Mathematics by Norman Priggs. Oxford.

152

9
GRAPH THEORY-II
Unit Structure :
9.0
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8

Objectives
Introduction
Walks, paths and circuits
Connected and disconnected graphs
Euler paths and circuits
Hamiltonian paths and circuits
Colouring of graphs
Let us sum up
References

9.0 OBJECTIVES :
After going through this chapter you will be able to:
find walks, paths and circuit in a graph
understand Eulerian graphs and circuits
understand Hamiltonian graphs and circuits
know about colouring the graphs

9.1 INTRODUCTION :
We have seen in earlier chapter, different operation of graphs
and different types of graphs. In this chapter we are actually perform
operation on graphs, so as to get circuit with a particular condition
imposed on the edges and vertices of a graph joining. The edges and
vertices in continuity we get a cycle or circuit. We can colour the
edges and vertices of graph with different colours and can also find
minimum number of colours required to colour a graph.

153

9.2 WALKS, PATHS AND CIRCUITS

Consider the following graphs.

G1

G2

G4

G3

G5

Fig. 9.1
In Fig. 9.1, observe graphs G1 and G 3 . We can travel from
one vertex to another just by traversing the edges. This is not
possible in case of the graphs G 2 , G 4 , G 5 as there is some vertex
from which the edge is not there to another vertex. At this stage we
are in position to make some more definitions.
Definition 9.1 :
A walk in a graph is defined as a finite alternating sequence of
vertices and edges, beginning and ending with vertices such that
each edge is incident with the vertices preceding and following it
and occurs exactly once.
Definition 9.2 :
The vertices at the beginning and at the ending of a walk are called
terminal vertices. Wherever terminal vertices are same, we call the
walk as a closed walk. A open walk is one which is not closed.
Definition 9.3 :
A open walk in which no vertex is repeated is called a path.
Definition 9.4 :
A closed walk in which no vertex is repeated is called a circuit or a
cycle.

154

Definition 9.5 :
The number of edges in a walk is called the length of the walk.
Let us illustrate all the definitions from 9.1 to 9.5 with help of graph
in Fig. 9.2.
e
1
v e3
1
e
2
v
2

e
6

e8

e
4
v6

v
3
e

e
10

e
5

9
v4

e
11
v
5

Fig. 9.2 A closed walk

An open walk :
v7 e3 v1 e1 v1 e2 v 2 e5 v5 e11 v 4 e10 v6 e4 v 2 . This is not a path as
vertex v1 and v 2 are repeated. The length of the walk is 7.

The closed walk :

v7 e7 v6 e8 v7 e6 v3 e9 v 4 e10 v6 e4 v 2 e2 v1 e3 v7 . The length of this
walk is 8. This is not a circuit as vertex v6 and v7 are repeated.

The path :
v5 e11 v 4 e10 v6 e7 v7 e6 v3 . Length of the path is 4.

The circuit or cycle :

v7 e7 v6 e10 v 4 e9 v3 e6 v7 . The length of cycle is 4.

Lemma 9.1:
Every uv walk contain uv path. u and v are terminal vertices.
Proof :The proof is by induction on the length of a uv walk say W.
Step 1:Suppose = 0 . It means the length of walk W is zero. It
means W contains single vertex u (=v). This vertex is u v
path of length zero.

155

Step 2:Suppose 1 and the result holds for walks of length less
than .
Step 3:If W has no repeated vertex then its vertices and edges form a
uv path. If W has repeated vertex w then delete all the edges
and vertices appearing between w. This gives a smaller uv
walk say W. W is contained in W.
As length of W is less than , induction hypothesis holds and
W is contained in W.
Theorem 9.2:
Let G be a graph with adjacency matrix A. V G = v1, v 2 , ..., v n
then the number of different paths of length r from vi to v j , where r
is positive integer equals to ijth entry of A n .
Example 1:
Find the number of paths from a to c for the graph in Fig. 8.3 of
length 3.
a

Fig. 9.3
Solution : Adjacency matrix A corresponding to graph in Fig. 9.3 is
a b c d e
a 0
b 0
A = c 0

d 1
e 1

0 0 1 1
0 1 1 1
1 0 0 1

1 0 0 0
1 1 0 0

0
1

then A3 = 2

4
5

We find A3 as we want to find path of length 3.

1 2 4 5
2 4 5 6
4 2 2 4

5 2 0 1
6 4 1 2

156

Number of paths between vertices a and c is the (1, 3)th entry of A3 .

(1, 3)th entry of A3 is two.

The paths of length 3 between the vertices a and c are two.

Check your progress
1) List all the different paths between vertices 5 and 6 in following
Figure 8.4. Give the length of each path.
1
a
c
4
e

2
d

Fig. 9.4
2) List atleast four cycles in graph given in Fig. 8.4.
3) Show that every uv-walk contains a uv-path.
4) Find the number of paths from vertex 1 to 3 for the graph in
Fig. 9.5 of length 3.
1

3
Fig. 9.5
5) Judge whether true or false.
1) Any uv-walk contains a uv-path.
2) The union of any two distinct uv-walks contains a circuit.
3) The union of any two distinct uv-path contains a circuit.

6) List all the paths that begin at vertex 2 in Fig. 8.4.

7) List all the circuits that begin at vertex 3 in Fig. 8.5.
8) In the following graph

157
2

Fig. 9.6

i) List all paths of length 2 starting from vertex 2 and 5.

ii) Find the cycle at vertex 7 & 4.
iii) List all paths of length 1.
iv) List all paths of length 3.

9.3 CONNECTED AND DISCONNECTED GRAPHS

A graph is connected if we can reach any vertex from any
other vertex by traversing along the edges. As seen in Fig. 8.1 G1
and G 3 are connected while G 2 , G 4 , G 5 are disconnected. A
disconnected graph consists of two or more connected graphs. Each
of connected parts in disconnected graph is the component of the
graph. More formally let us give definition of all these.
Definition 9.6:
A graph G with u, v V(G) is said to be connected if G has a
uv path otherwise G is said to be disconnected.
Definition 9.7 :
A disconnected graph is not a single piece. Each single piece
in a disconnected graph is called a component. A component is
maximal connected subgraph of given disconnected graph. For
example, in Fig.9.1, G 2 has two components.
H2 :

H1 :

and
G 4 has two components say K1 and K 2

158

K1 :

K2 :

G 5 has five components each having a single vertex.

Theorem 9.3:
A graph G is connected if and only if given any pair u, v of
distinct vertices, there is a path from u to v in G.
Proof :Let G be disconnected. Then G is expressed as a disjoint
union of subgraphs. Choose two vertices u and v which
belong to two of these different subgraphs say G1 and G 2 .
Since G is union of mutually disjoint subgraphs, we cannot
find an edge which is incident to vertices in G1 and G 2 .
Hence we cannot find a path from u to v. This is a
contradiction. G, therefore cannot be expressed as union of
mutually disjoint subgraphs.
G is connected.
Conversely, assume that G is connected and there are two
vertices in G such that there is no path between them. Let u
& v be these vertices. Denote by G1 the induced subgraph
formed by all those vertices w of G such that there is uw path
in G. Let G 2 denote the complement of G1 . Then G is union
of these two mutually disjoint subgraphs i.e. G is
disconnected. This is contradiction. Hence we cannot find a
pair of vertices such that there is no path between them.
Theorem 9.4:
Let G be a simple graph on n vertices. If G has k
components then the number q of edges of G satisfies the inequality,
1
2

n - k q n - k n - k +1
Check your progress
1) Draw complete graph on seven vertices. Is this graph connected?
2) Give an example of a regular, connected graph on six vertices
which is not complete.

159

3) Construct a graph with 8 vertices 16 edges and degree of each

vertex 4.
4) Judge whether the following are true or false.
i) A graph G is connected if and only if it has exactly one
component.
ii) Complement of connected graph is connected.
iii) Subgraph of a connected graph is also connected.
iv) If G is graph on 5 vertices and G has two components then
number of edges of G is atmost 12.
5) Let u and v be any two vertices of a connected graph G. Show
that there exists a uv walk containing all vertices of G.
6) Draw a simple (p, q) graph G to show that G contains unique
circuit iff G is connected & p = q.
7) Which connected graphs can be both regular and bipartite?
8) Give an example of regular connected graph on six vertices that
is not complete.
9) Give an example of a graph on five vertices with exactly two
components.
10)Give an example of a graph that is regular but not complete, with
each vertex having degree three.
11)Give an example of a graph with seven vertices and exactly three
components.

9.4 EULER GRAPHS :

In this section we consider the task to travel a path using each
edge of the graph exactly once. The problem is to draw a figure
without lifting the pencil from the paper and without retracing a line.
Definition 9.8:
A path containing all the edges in a graph is called an
Eulerian path.
Example 2 :
The figure 8.4.1 shows examples of Eulerian graphs and non
Eulerian graphs.

160
v
1

v
2

12

10

v
5

v
4

v
3

v
2

10

v
8

11

e3

v
3

v
1

v
6

v
5

v
4

v
6

v
7

8
7

(a) Eulerian graphs

(b) Non Eulerian Graphs

Fig. 9.4.1
Definition 9.9:
A closed walk containing all the edges in a graph is called an
Eulerian circuit.
A graph is Eulerian graph, if it has a closed walk containing
all the edges.
In Fig. 9.4.1(a) graph G1 is Eulerian because it contains Eulerian
circuit v1 e1 v2 e2 v3 e3 v4 e4 v5 e5 v6 e6 v1 e7 v5 e8 v2 e9 v4 e10 v1 .
In Fig. 9.4.1(a) graph G 2 is Eulerian because it contains Eulerian
circuit
v1 e1 v2 e2 v3 e3 v4 e4 v2 e5 v5 e6 v6 e7 v7 e8 v4 e9 v3 e10 v7
e11 v8 e12 v1 .
Lemma 9.5:
If every vertex of a graph G has degree atleast 2, then G
contains a cycle.

161

Theorem 9.6: Characterisation of Eulerian graph :

A connected graph G is Eulerian if and only if every vertex of
G is of even degree.
Corollary 9.7:
Let G be connected with exactly two odd vertices say u and v.
then there is a uv walk in G that contains all edges of G.
Proof :Let G be the graph obtained from G, by adding an edge
joining u and v. The degree of each vertex in G is even and
therefore it has Eulerian walk beginning with u and ending
with u. Any Eulerian walk for G must contain the edge uv.
Otherwise, G itself would be Eulerian and u as well as v will
be of even degree which is a contradiction. Let therefore,
v u v1 v 2 ..... vi v, be Eulerian walk of G .
Then the walk u v1 v2 .... vi v got by removing edge
uv is required uv walk in G.
Remark: As mentioned in 7.2 about application of graphs, the
Konigsberg bridge problem was to find an Eulerian walk in
Eulerian graph. Euler proved that it is impossible to get
Eulerian walk in the given graph. Therefore, there is no
solution to Konigsberg bridge problem.

Next, we give an algorithm that produces an Euler circuit for

a connected graph with no vertices of odd degree.
Definition 9.10:
An edge is called a bridge or cut edge in a connected graph
if deleting it would create a disconnected graph.
See Fig. 8.4 for an example of a bridge. The edge e5 is a
bridge. Remove e5 to get disconnected graph G - e5 .

162
v2
e1
v
1

e2 e
5

v8

e6

v3
e3

e4

v2

v6

v7

v4

v6

v8

v
1

e8

v3
v7

e7

v4

G - e5

Fig. 9.4.2 A bridge or cut-edge

Algorithm 9.11 : Fleurys algorithm
Let G be an Eulerian graph and v be any vertex of G. Starting
from v, we traverse the edges of G subject to following conditions
1. Erase an edge that is traversed
2. Erase the isolated vertices, if any.
3. Traverse a bridge only if there is no other alternative.
The procedure is possible and finally we arrive at an empty
set. The successive sequence of edges which were removed will
form an Eulerian walk for G.
Example 3: To illustrate Fleurys algorithm.
Use Fleurys algorithm to construct Euler circuit for the graph
in Fig. 9.4.3.

Fig. 9.4.3
Start with vertex v.
Step 1 : Traverse any edge incident at v say e1

163

Step 2 : Erase the edge that is traversed. The graph will now look
like this
v

Step 3 : Traverse the edge e2 .

Step 4 : Erase edge e2 the graph will now be
v

e3
v

e8

e9
v

At v5 we have three choices for edges viz, e3 , e8 , e9 . But edge

e9 is bridge. Thus, by condition 3 of the algorithm we cannot
traverse this edge.
Step 5 : Traverse edge e3 .
Step 6 : Erase edge e3 . The graph will be
v

Step 7 : Traverse e4 and erase e4 to get graph.

164

Though e4 was bridge but we had no alternative.

Step 8 : Erase vertex v 4 .
Step 9 : Traverse e5 and erase e5 to get the graph.
v
3

v
5

v
1

e v
6 2

Step 10 : Traverse e6 and erase e6 to get the graph. Also remove

vertex v 2 .

v
3

v
5

e
7
v

v
1

Step 11 : Traverse e7 , erase e7 and remove v1 to get,

v
5

v
3

v
Step 12 : Traverse e8 , erase e8 and remove v3 to get
v

165

Step 13 :
Traverse e9 , erase e9 and remove v5 . The graph left
is single vertex.
Step 14 : Remove v v .
The

sequence

of

edges

erased

in

order

were

e1 , e 2 , e3 , e 4 , e5 , e 6 ,
e7 , e8 , e9 .
Thus
we
get
Eulerian
circuit
v v1 v5 v 4 v3 v 2 v1 v3 v5 v observe that

each edge of G is traversed exactly once. However, the vertex may

be traversed more than once. The graph should remain connected at
every step.
Check your progress
1) Make adjacency and incidence matrix of any one Eulerian graph.
What can you say about them?
2) Draw a Eulerian graph on 6-vertices.
3) Which of the following graphs have an Eulerian circuit, an
Eulerian path but not an Eulerian circuit or neither? Give reason
for your choice.

a)

b)

c)

d)

e)

166

4) Use Fleurys algorithm to find an Eulerian path for the following

graphs :

a)

c)

b)

d)

e)

5) Judge whether the following are true or false.

i) Eulerian graph is connected.
ii) K n is Eulerian if and only if n is even.
iii) K m,n is Eulerian if and if m and n are even.
iv) Line graph of Eulerian graph is Eulerian.
v) Petersens graph is Eulerian graph.

9.5 HAMILTONIAN GRAPHS :

We have seen the walks which include all the edges of the
graph. Now we see the graph which contains paths that contain each
vertex of the graph. These graphs were named after Sir William
Rowan Hamilton who introduced such graphs.

167

Definition 9.11:
A cycle which passes through each vertex of the graph called
Hamiltonian cycle or circuit. We say Hamiltonian cycle spans the
graph. A Hamiltonian path is a path that contains each vertex
exactly once. A graph having a Hamiltonian cycle is called
Hamiltonian graph.
Note : Hamiltonian graph contains a closed path that includes all the
vertices of the graph.
Example 4 :
Consider the graphs in Fig. 8.5.1.
v
1

u
1

v
2
u

v
6

v
3
v
5

v
4

2
u
3

u
4

(a)

(b)
Fig. 9.5

The graph of Fig. 9.9.1(a) is Hamiltonian graph.

Hamiltonian cycle is v1 v 2 v3 v 4 v5 v6 v1 .

The

The graph of Fig. 9.9.1(b) is not Hamiltonian graph.

However, the walk v6 v3 v 4 v5 v1 v 2 is the path that
includes all the vertices but it is not closed.
Example 5 :
Any complete graph K n has Hamiltonian cycle.
Hamiltonian graph.

Hence K n is

Note : If G is graph on n vertices and G has Hamiltonian circuit then

G must have atleast n edges.
Remark :1)It is not always possible to determine a Hamiltonian
path or cycle.

168

2)If there is Hamiltonian path or circuit in a graph then

there is no efficient way found to trace it, unlike,
Eulerian path or cycle.
Some observations of Hamiltonian graphs are as follows :

Theorem 9.8 :
If G is Hamiltonian then for every non empty proper subset S
of V(G), W(G S) |S| where W(H) denotes number of components
of any graph H and |S| denotes number of elements of S.
Proof :Let C be Hamiltonian cycle of G. If S is any proper non
empty subset of V(C) = V(G). If G = C, we are through.
Otherwise G can only be C with more edges. But the addition
of edges to C can only decrease the number of components of
C S.
Therefore, W(G S) W(C S)
i.e. W(G S) | S |
Remark : The above theorem is useful in showing that some graphs
are non-Hamiltonian. For example, the complete bipartite
graph K m,n where m n is non-Hamiltonian. Let V1, V2 be
partition of vertex set Km,n where V1 =m . The graph
Km,n V1 is totally disconnected with n vertices as its
components. Thus, W Km, n V1 = n m =| V1 |
Note : The converse of theorem 7.9.4 is not true. For example, the
Petersens graph satisfies the condition of the theorem but is
not Hamiltonian.
Theorem 9.9 :
A Hamilton graph contains no cut-vertices.
Theorem 9.10:
Let G be simple graph with n vertices. Suppose a and v is a
pair of non adjacent vertices such that deg.(v) + deg.(u) n . Then G
is Hamiltonian. The proof of above two theorems is omitted but
from it we can prove the following :

169

Corollary 9.11 :
G has Hamiltonian cycle, if each vertex has degree greater
than or equal to n 2 .
Proof :The sum of the degrees of any two vertices of G is
n + n = n . Thus, all conditions of theorem 7.9.6 are
2
2
satisfied.
Hence, G has Hamiltonian cycle i.e. G is
Hamiltonian graph.
Theorem 9.12 :
Let G be a graph with n-edges. Then G has Hamiltonian
cycle if m

1 2
n 3n + 6 where n is number of vertices of G.
2

Proof :Suppose u and v are non-adjacent vertices of G.

Let H be the graph obtained by deleting vertices u and v
along with any edges having u or v as end point. Then H has
n 2 vertices. The number of edges of H is m deg(u) deg
(v). [If u and v were adjacent then one edge less will be
removed] The maximum number of edges H could possibly
have is n 2 C2 . This happens if there is an edge connecting
every distinct pair of vertices.
Thus, the number of edges of H is atmost
n 2

C2 =

n 2 !

2! n 2 2 !

n 2

n 2 !
2! n 4 !

n 2 n 3

2
1 2
= n 5n + 6
2

1
2
1
deg(u) + deg(v) m (n 2 5n + 6)
2

Therefore, m deg(u) deg(v) (n 2 5n + 6)

1
2

Given that m (n 2 3n + 6)

C2 .

170
1
2

1
2

Therefore, deg(u) + deg(v) (n 2 3n + 6) (n 2 5n + 6)

1 2
(n 3n + 6 n 2 + 5n 6)
2
1
= (2n)
2

=n
The condition of theorem 9.10 is satisfied and G has
Hamiltonian cycle.
Remark : The converse of theorem 9.10 and theorem 8.12 is not
true. Example, consider the graph G in Fig. 9.5.2. |V(G)| = n
= 8. Each vertex has degree 2. If u and v are vertices of G
then deg(u) + deg(v) = 4 .

Fig. 9.5.2
The total number of edges is 8. The conditions in the theorem
is not satisfied still we can find a Hamiltonian cycle in G.
Definition 9.12 :
A graph G is called a weighted graph if there is a positive real
number associated with each edge of G. The real number is called
the weight of the corresponding edge. The weight of subgraph H of
a weighted graph is defined as the sum of weights of all edges of H.
Travelling Salesman Problem :
As mentioned in 9.2 about the application of graph, the
traveling salesman problem is to find a minimum weight
Hamiltonian cycle in a weighted Hamiltonian graph.
No efficient algorithm for solving the traveling salesman
problem is known. However, it is possible to obtain a reasonably
good but not necessarily optimal solution. One such method is given
below. First find Hamiltonian cycle G. Then search for another
Hamiltonian cycle of smaller weight as follows :

171

Fig. 9.5.3
Let u, v and x, y be two non adjacent edges in C such that
the vertices u, v, x and y occur in that order in C. See Fig. 8.5.3 If
u, x and v, y are edges and w u, x + w y, v < w u, v + w x, y
then replace edges (u,v) and x, y in C by edges u, x and (v,y).
The resulting cycle C would be Hamiltonian and will be of less
weight than C. continue above process with C , until one gets
reasonably good Hamiltonian cycle.
Note : The initial Hamiltonian cycle itself may be chosen such that it
has comparatively small weight.
Example 6 :
Find the Hamiltonian cycle of minimal weight for the following
graph 9.5.4.
B
12
16

20

15

10
C

D
16

12

15

12

14
F

10

Fig. 9.5.4 Hamiltonian circuit

Consider vertex A vertex. Vertices B, G and D are adjacent to
A. w(A,B)=12, w(A,D) = 20 and w(A,G)=12. We have to choose
minimum weight. Let us consider edge AB with weight 12. As we
reach B there are two choices either C or D. Min {w(B,C), w(B,D)}
= w(B,C)=10. So we reach C with w(B,C) = 10. C is adjacent to F
and w(C,F) = 15. At F there are three choices either B, G or E.

172

Min {w,(F,B), w(F,G), w(F,E)}

= w(F,E) = 10.
Thus, we take vertex E. So far we have traversed as follows :
12

10

15

10

A B C F E . From E we reach D, w(E,D) =

12. From D we have to go to G, as B is already traversed. From G

we reach A back again. Thus, the Hamiltonian cycle of minimal
weight for the graph in Fig 8.5.4 is
A

12

12

10

15

16

F
D

10

12

The weight of this cycle is 87 as

w(AB) + w(BC) + w(CF) + w(FE) + w(ED) + w(DG) + w(GA)
= 12+10+15+10+12+16+12
= 87
Check your progress
1) Give two Hamiltonian circuits in K5 that have no edges in
common.
2) Give example of Hamiltonian graph which is not Eulerian.
3) How many distinct Hamiltonian cycles are there in K 4 and K 6 ?
4) Show that there are only three Hamiltonian graphs on 6-vertices,
C6 , K 6 and K 3,3 .
5) Give one example of an Eulerian graph which is not
Hamiltonian.
6) Justify whether true or false.
i) Every Hamiltonian graph is connected.
ii) An induced subgraph of Hamiltonian graph is Hamiltonian.
iii) A simple graph which is both Hamiltonian and Eulerian is
necessarily a cycle.

173

7) Which of the following graphs have Hamiltonian circuit, a

Hamiltonian path but no Hamiltonian circuit, or neither. Trace
the circuit if graph has Hamiltonian circuit.

a)

b)

c)

d)

e)

f)

8) Find a minimal weight Hamiltonian circuit for the graphs given

below.
a)

b)

2
C

2
6

C
5

5
G

6
3

4
F

174
c)

B
2

5
4
I

1
2

1
G

3
F

D
2
E

9.6 COLOURING OF GRAPHS :

Definition 9.13 :
Consider a graph G. A vertex colouring or colouring of G is
an assignment of colours to the vertices of G such that adjacent
vertices have different colours.
A graph is said to be k-colourable if the vertices of G are
coloured using atmost k-colours such that adjacent vertices receive
different colours.
The chromatic number denoted by (G) of a graph G is the
minimum number of colours needed to colour G.
Note : If a graph is k-colourable then it is (k+1) colourable.
Examples :
1) Kp is p-colourable..As each pair of vertices is adjacent.
Kp = p . Consider K 4 in Fig. 8.6.1

Fig. 9.6.1 Chromatic number of K 4 is 4.

K4 = 4

2) Kp is complement of Kp .
components as vertices.

It is disconnected graph with p

Kp is 1-colourable Kp = 1 .

shown in Fig. 8.6.2, K 4 is 1-colourable and K 4 = 1 .

As

175

Fig. 9.6.2 Chromatic number of K 4 is one.

3) Consider bipartite graph K m,n (V1,V2 ) with m-vertices in V1 and
n-vertices in V2 . No two vertices of V1 and V2 are adjacent.
But every vertex of V1 is adjacent to each vertex of V2 .
Therefore, every vertex of V1 can be coloured by colour and
all vertices of V2 can be coloured by colour . Thus

K m,n = 2 . Consider K 2,3 as in Fig. 9.6.3.

(a )

u ( b)
1

v
2

u ( b)
2

(a)

3 (b)

Fig. 9.6.3 : Colouring of K 2,3

V1 = v1,v 2 ;V2 = u1,u 2 ,u3 , Chromatic number of K 2,3 is two.

Theorem 9.13 : Graph G is 2-colourable iff G is bipartite

4) Consider the graph G in Fig. 9.6.4.
v
1 (a)
(b )
v

(9 )

v5

v2 (b)

v3 ( w)

Fig. 9.6.4
Let vertex v1 be coloured by colour . Then v 4 and v5
must receive two distinct colours say and 9 different from .
Now, vertex v3 is adjacent to v1 , v4 and v5 . Therefore, it cannot

176

have colour , ,9 . Hence v3 will receive colour . The vertex

which is to be coloured is v 2 . v 2 can receive colour as it is not
adjacent to v 4 . Thus, atleast 4 colours are necessary to colour the
graph. Hence (G) = 4 .
Definition 9.13 :
A graph is called planar graph if it can be drawn in a plane so
that no two edges of graph cross.
The crossing is said to occur in a graph it two edges meet in a
point which is not a vertex.
The figure 9.6.5 shows an example of planar and non-planar
graph.
v
1

v
4

v
2

v3

(a) Non-planar Graph

v
1

v
4

v
2

v3

(b) Planar Graph

Fig. 9.6.5
In Fig. 9.6.5(a) edges v 2 v 4 and v1v3 cross each other. In Fig.
9.6.5(b) two edges wherever they meet they meet only at their end
points. The graphs (a) and (b) are isomorphic.
Theorem 14 : Every planar graph is 6-colourable.
Theorem 15 : (Heawood) Every planar graph is 5-colourable.
Theorem 16 : (four colour theorem) Every planar graph is 4colourable.
Note : The proof of theorem 4 was given by computer computations.
A computer free proof of this theorem is still to be found.
Appel and Haken have solved four colour theorem by using
computers.
Application of colouring of graph :
The colouring of graphs were used to solve map-colouring
problem. The colouring of map means to colour each region (or
country or state or province) so that no two distinct regions sharing a

177

common border have the same colour. The map-colouring problem

is to find the smallest number of colours used to colour the map.
Each region will be considered as vertex and an edge will be
common boundary shared by two regions. For example, consider
the map given in Fig. 7.10.6 and its graphical representation with
colouring.

UT

NE
UT

(a)

CO
KS

AZ

NM

(v)
AZ

(a) Map G

CO

(b)

NE
(9 )

( a) KS

(9 )

NM

(b) Graph of map G

G = 4

Fig. 9.6.6
Definition 9.14 :
Given an integer > 0 , we denote by PG () as the number of
different - colourings of graph G. We call PG () as chromatic
polynomial of G.
Example :
1) Consider simple trivial graph K1 . K1 is graph with one vertex.
We can colour K1 with any of given colour. Therefore,
PK 1 () = .
2) Consider graph K 2 , a graph with two incident vertices u and v.
We can colour vertex u with colours but v will carry the colour
which is not given to u. So total colourings of v are ( - 1)
colourings.
Thus, total colourings of K 2 given by
PK 2 () = ( 1) .
3) In General, PK n () = ( 1) --------- ( n + 1) .
4) Consider the line graph L4 as shown in Fig.9.6.7.
v1

v
2

v
3

v
4

Fig. 9.6.7 Line graph L 4

178

Suppose there are - colours. The first vertex can be coloured

by any colours. The second vertex adjacent to first can be
coloured by any ( 1) - colours. 3rd vertex can be coloured by any
th
( 1) colours not used for second vertex. Similarly, the 4 vertex
can be coloured by any ( 1) - colours. By multiplication principle
of counting, the total number of colourings is ( 1)3 . Thus,
PL4 = ( 1)3 .

Theorem 9.17:
Let G1, G 2 ,...., G K be components of disconnected graph G.
Then PG () = PG1 ().PG 2 (),...., PG K () , where R.H.S. is product of
chromatic polynomials of each component.
Example:
Consider graph G in Fig.9.10.8. It has two components each of
which is K3 . The chromatic polynomial of K3 is 1 2 .

PG () = 2 12 22

Fig. 9.6.8
As (G) = 3 , the number of distinct ways to colour G using 3
colours is PG (3) = 32 3 12 3 2 2
= 9(2) 2 (1)2
= 9 4
= 36

179

Check your progress

1) Construct a graph for the following maps.

a)

b)
P

T
S

c)

d)

2) Find the chromatic number for the following graphs.

b)

a)

c)

d)

e)

3) Find the chromatic polynomial PG for the graphs given in

Exercise 2.
4) Determine whether the given graphs are planar? If so, draw it so
that no edges cross.

a)

b)

180

c)

d)

5) Find PG and X(G) for the graph drawn in Exercise 1.

6) Give example of a connected graph on five vertices that is (a)
planar (b) not planar.
7) What is X(G) for a bipartite graph G? Justify.
8) Judge whether following is true or false.
i) The chromatic number of any cycle is 2.
ii) The chromatic number of bipartite graph is 2.
iii) For each integer p > 0 there is a graph which is p-colourable.
iv) The chromatic number of Petersens graph is 4.
v) Any two graphs with same chromatic number are isomorphic.

9.7 LET US SUM UP :

In unit 8, we have seen interesting graphs like Eulerian graphs
and Hamiltonian graphs. We have also leant simple methods to find
Eulerian circuit and Hamiltonian circuit. As colouring of graphs also
is a field where not much work is done, we have tried to learn basic
concepts of colouring a graphs.

9.8 REFERENCES
1. Discrete Mathematical structures by Kolamn, Busby and Ross.
Pearson education.
2. Introduction to graph theory by Douglas B. West.
3. Discrete Mathematics and its applications by Kenneth. H.Rosen.
McGraw Hill edition.
4. Graph theory by Frank Harary. Narosa Publication.
5. Discrete Mathematics by Norman Priggs. Oxford.

181

10
TREES
Unit Structure
10.0 Objectives
10.1 Introduction
10.2 Characterisation of Trees
10.3 Labeled Trees
10.4 Minimal Spanning Trees
10.5 Trees Searching
10.6 Directed and Undirected Trees
10.7 Unit End Exercises
10.8 Let us sum up
10.9 References for further reading

10.0 OBJECTIVES
After going through this chapter students will be able to
understand:
Know about tree in graph theory, its properties and
characterisation
Find minimal spanning trees
Know about directed and undirected trees

10.1 INTRODUCTION :
One of the most important topic in study of Graph Theory is
Trees. Many applications rangeing from family tree to computer
science management tree is dealt with the study of trees.
The first use of trees by German Mathematician Karl Georg
Christian von Staudt was done in 1847 in his work on projective
Geometry and by German physicist Gustav Robert Kirchhoff in the
same year in his paper on electrical networks. The word tree for a
particular graph was first used by Arthur Cayley.

182

In this unit we are going to study trees and its properties first.
Then we study labeled trees; spanning trees, rooted and binary trees.
The wide range of application of trees in different field has created a
revolution in study of trees.
A graph of an undirected tree T will have a single line
without arrows. While the edge of directed tree will have arrows
showing the direction of edge.

10.2 CHARACTERISATION OF TREES :

To start with we have to first understand basic definitions
related to trees.
10.2.1 Definition : A connected graph which contains no circuits is
called a tree.
We can also frame a definition which uses set theory notation.
Let A be a set and T be relation on A then we say that T is a tree if
there is a vertex v0 in A with the property that there exists a unique
path in T from v0 to every other vertex in A but no path from v0
to v0 .
Remark : Tree is a connected acyclic graph.
Some examples of trees is given in figure 10.2.1.

(a)

(b)

(c)

(d)

Fig. 10.2.1
10.2.2 Properties of Trees
Let T be a graph on n vertices then
1. There is one and only one path between every pair of vertices in
a tree.
2. T has n vertices and n 1 edges.
3. T is connected and has n 1 edges.

183

4. T is connected and every edge is cut-edge.

5. T is minimally connected graph.
6. T has n-vertices, n 1 edges and no circuits.
7. In a tree with n 2 , there are atleast two vertices of degree one.
10.2.3 Definition : A pendant vertex is defined as a vertex of degree
one.
Example : In the graph given below in Fig. 9.2.2 v0 is the pendant
vertex. There are more than one, pendant vertex in any tree.

v0

Fig. 10.2.2
10.2.4 Definition : The initial vertex v0 considered when we start
drawing a tree is called root vertex. A tree in which there is root
vertex is called rooted tree. In rooted tree the root vertex is clearly
distinguished from remaining vertices. Denote tree T with root
vertex v0 as T,v0 . Examples of rooted trees are given in Fig.
10.2.3.

Fig. 10.2.3 Rooted trees

10.2.5 Definition : A binary tree is defined as a tree in which there
is exactly one vertex of degree two and each of remaining vertices
are of degree one or three.

184

Examples of binary trees are given in Fig. 9.2.4.

Fig. 10.2.4 : Binary Trees

Note that every binary tree is a rooted tree.
10.2.6 Theorem 1 : Let T be a tree with root vertex v0 then,
a) There are no cycles in T.
b) v0 is the only root of T.
c) Each vertex in T, other than v0 has in-degree one and v0
has in-degree zero.
10.2.7 Levels in a tree
Consider the trees given below in the figure 10.2.5. We start
with vertex v0 the root vertex. No edges enter v0 .
Level
v0

v
4
v8

v
3

v2

v1

v0

v
6

v5

v9

(a) T1

v
10

v
7
v
11

v2

v1
v
4

v
3

v6

v5

v7

(b) T2
Fig. 10.2.5 : Levels in a tree

Several edges may leave v0 . These edges are drawn in

downward direction. The vertex v0 is said to be at level 0. The
edges starting at v0 will terminate into the vertices which will be
called as level 1 vertices. v0 is sometimes called parent of vertices

185

in level 1 while vertices in level 1 are called the offspring of v0 .

The edges having a vertex at level 1 are drawn downwards. The
vertices at which the edges drawn from level 1 vertices terminates
are called vertices at level 2. The offspring of any one vertex are
sometimes called siblings. The largest level number of a tree is
called the height of the tree. The vertices of the tree that have no
offspring are called the leaves of the tree. All the vertices of tree
that can be reached by a path beginning at any vertex v are called
descendants of vertex v. If T is rooted tree with vertex v0 and v is
any vertex of T other than v0 then the tree with root vertex v is
called subtree of tree T beginning at v. Denote subtree of tree
(T, v0 ) as T(v).
In figure 10.2.5(a) T1 has vertex v0 at level 0. Level 1
vertices are v1, v 2 , v3 . Level 2 vertices are v 4 , v5 , v6 , v7 . Level 3
vertices are v8 , v9 , v10 , v11 . The height of tree T1 is 3. The vertices
v8 , v9 , v10 , v11 are leaves of tree T1 . The offsprings of v0 are
v1, v 2 , v3 . The offspring of v 2 are v5 and v6 . The vertices v5 and
v6 are siblings of v 2 . The descendants of v 2 are v10 and v11 . The
subtree of T1 beginning at v 2 is given in figure 10.2.6. The height
of subtree is 2.
Level
v2

0
v6

v5

10

11

Fig. 10.2.6 : Subtree of T1

Example

Draw

tree

with

vertex

set

{v1 , v 2 , v3 , v 4 , v5 , v 6 , v 7 , v8 , v9 ,

v10 } and edge set { v2 , v3 , v2 , v1 , v 4 , v5 , v 4 , v6 , v5 , v8 , v6 , v7 ,

v4 , v2 , v7 , v9 , v7 , v10 } . Show that T is a tree and identify the

root. Draw one subtree of T. Write height of T.
Solution : T is a tree with root vertex v 4 as there is path from v 4 to
every vertex in tree T. There are no cycles. Height of tree T is 3.

186

Level

v
4

0
1

2
3

v2

v
6
v7
v
9

v
5
v
3

v1

v8

v
10

(a) T
Subtree of T is

v6
v7
v9

v10

(b)
Fig. 10.2.7 Subtree of T

10.3 LABELED TREES

It is sometimes useful to abel the vertices of tree for a
particular purpose. The set of vertices of tree is not important but
emphasis is given on the label which is attached to vertices of tree.
Such a tree where labels are given to each vertex is called labeled
tree.
Some examples of labeled trees.
10.3.1 Label tree representing algebraic expression.
Consider the algebraic expression.

4 3 x + 2x + 5+x

187

The labeled tree for this expression is shown in Fig. 10.3.1.

+
x

Fig. 10.3.1 : Labeled tree for algebraic expression

While drawing such tree, we try to find the central operator.
The central operator is the operation which is inbetween two
numbers or expressions. As shown in the example. + is central
operator of expression 4 3 x and 2 x + 5 + x . is
central operator of number 4 and expression 3 x while + is
central operator of expression 2 x and 5 + x . Similarly, - is
central operator of 2 and x ; + is central operator of 5 and x ; x is
central operator of 3 and x . Each vertex has only one label either a
operation sign or the number.
10.3.2 Positional tree : It is a type of labeled tree. The vertices are
labeled so as to show the position of offspring. While drawing a
positional tree we have to imagine that the n offspring positions for
each vertex are arranged symmetrically below the vertex and we
place in its appropriate position each offspring that actually occurs.
We do not label the root vertex in these trees as root vertex is
not a offspring. Suppose we choose three offspring positions then
each vertex will have offsprings placed at these 3 positions only.
If any offspring is not present then that position will not be shown.

188

3
1

Fig. 10.3.2 A positional 3 - tree

As shown in figure 10.3.2 the first offspring 1 from any
vertex v if it exists is edge drawn from v sloping to its left. The
offspring 2 from vertex v if it exists is drawn vertically downward
from v while offspring 3 from vertex v if it exists, is drawn to the
right of vertex v dropping downloads.
10.3.3 Labeled graph : A graph in which every vertex is assigned a
unique name or label (no two vertices have the same label) is called
a labeled graph.
Whenever we are counting the total number of distinct
labeled graphs on certain number of vertices the two differently

labeled graphs are counted separately even though they are

isomorphic. Note that the number of distinct labeled trees with n
vertices n 2 is n n 2 . For example, consider labeling of trees on
4 vertices. There are 16 trees on 4 vertices that carry distinct
labels. Some of these labeled trees on 4 vertices are shown in Fig.
9.3.3.
A

BA

Fig. 10.3.3 : Labeled trees

The trees in figure 10.3.3 are isomorphic but are different
labeled trees.

189

9.4 MINIMAL SPANNING TREES :

Here we see how are can associate tree to a connected graph.
Definition 10.4.1 : Let G be a connected graph then a spanning tree
T of G is a subgraph of G which is a tree and which contains all the
vertices of G.
Example : Consider graph G and its spanning tree in figure
10.4.1(a) and 10.4.1(b) respectively.

a)

v7

v3

v2

v8

v4
v

v2

v7

b)

v8
v

v4
v

6
Fig. 10.4.1 : Spanning tree

We remove the edges of a graph G which forms cycle and get

a tree.
Definition 10.4.2 : A minimal spanning tree in a connected
weighted graph is a spanning tree that has the smallest possible sum
of weight of its edges.
There are two algorithm for constructing minimal spanning
trees. Both proceed by successively adding edges of smallest
weights, from those edges with a specified property that have not
already been used.

190

10.4.3 Prims algorithm : Let G be a graph with n-vertices.

Step 1 Choose a vertex v1 of G. Let V = v1 and E =

Step 2 Choose a nearest neighbour vi of V that is adjacent to

v j , v j V and for which the edge vi , v j does not form a cycle

with members of E. Add v j to v and edge vi v j to E.

Step 3 Repeat step - 2 until |E| = n 1 , Then v contains all n

vertices of G and E contains all the edges of minimal spanning
tree T.
Example : The small town of social circle maintains a system of
walking trails between the recreational areas in town. The system
for the same is shown in Fig. 10.4.2. The system is a weighted
graph. The weights represent the distances in kilometers between
sites.

v0

v2

2
5

v7

v4

4
4

Fig. 10.4.2
Using Prims algorithm we begin with vertex v0 . The nearest
neighbour to v0 is v 2 who is 2 km away from v0 . So we select
edge v0 v 2 first. Fig. 9.4.2(a). Consider set { v0 , v 2 }. The vertex
v1 is the nearest neighbour. We can either choose v0 v1 or v 2 v1 as
next edge. Let us choose v 2 v1 . Then vertex set will be { v0 , v 2 , v1 }
and edges set will be { v0 v 2 , v 2 v1 }. Fig. 9.4.2(b). Here we cannot
choose edge v0 v1 as it will make a cycle so next vertex we choose
is v5 and edge v 2 v5 to get { v0 , v 2 , v1 , v5 } and edge set
{ v0 , v 2 , v 2 v1 , v 2 v5 }. Continue in this manner and finally get
minimal spanning tree shown Fig. 10.4.2.

191

v0

v0

(a)

v2

v0

(b)

(c)

v0

3 2

v2

v4

5
4

(d)

Fig. 10.4.2: Minimal spanning tree using Prims algorithm

Thus total weight of the spanning tree which is minimal is
2 + 3 + 5 + 4 + 2 + 3 + 2 = 21. Thus, the bicycle path of minimum
length for the system is of 21 km.
Note that for the same graph shown in Fig. 10.4.2. We can
find minimal spanning tree using Prims algorithm beginning with
vertex v 4 . Here also bicycle path is 21 km, long as seen in Fig.
10.4.3.
v
v
1
3 2
v
3

v0

5
2

v4

v2

Fig. 10.4.3 Minimal Spanning Tree

10.4.4 Kruskals algorithm : Let G be a connected graph with n
vertices and let S = { e1, e2 ,....., ek } be set of weighted edges of G.
Step 1 Choose edge e1 in S of least weight. Let E = { e1 }. Replace S
with S{ ei }.
Step 2 Select edge ei in S of least weight that will not make a cycle
with edges of E. Replace E with E ei and S with S{ ei }.
Step 3 Repeat step - 2 until |E| = n 1.
Since G has n vertices, the n 1 edges in E will give spanning
tree T.

192

Example : Consider the graph given in Fig. 10.4.3.

12
20

v0

v2

10
16

15

15

12

20

12

14

10

v4

Fig. 10.4.3
Initially choose edges with minimum weight. In Fig. 10.4.3,
they are edges v1 v 2 and v5 v 4 . Both of these are selected. Fig.
10.4.4 (a) Next there are three edges of weight 12. All these edges
can be added without creating a cycle. Fig. 10.4.4 (b) Edge of weight
14 is remaining edge of least weight. Adding this edge given us six
edges for 7 vertex graph, so a minimal spanning tree is found as
shown in Fig. 10.4.4 (c).
v0

v2

10

12

v2

10

12

10

12

v4

v4

10

(a)

(b)
v0

v2

10

12
12

6
14

12

10

v4

(c)
Fig. 10.4.4 Minimum spanning tree by Kruskals algorithm

193

10.4.5 Application of minimal spanning tree

Suppose that n cities v1, v 2 ,...., vn are to be connected through
a network of roads. The cost Cij of building a direct road between vi
and v j is given. Then the problem is to find the least expensive
network that connects all n cities together. Thus the problem of
connecting n cities with a least expensive network is a problem of
finding a shortest spanning tree in a connected weighted graph of n
vertices.

10.5 TREES SEARCHING :

Sometimes it is necessary to consider each vertex of a tree T
exactly once in some specific order. When we reach a particular
vertex and wish to perform computation at that vertex then that
application is represented by the tree. By visiting a vertex, we mean
performing particular task at that vertex.
The process of visiting each vertex of a tree in some specific
order is called searching the tree or tree search.
In this topic we will consider searches on binary positional
trees. We know that in a binary positional tree, each vertex has two
offsprings. Denote these offspring as v L and v R where v L denotes
left offspring and v R denotes right offspring. If a binary tree is not
positional then it can be labeled in such a way that it becomes
positional.
10.5.1 Definition : Let T be positional binary tree with root vertex v
then if v L exists then the subtree T(v L ) will be called the left subtree
of T and if v R exists then the subtree T( v R ) is called right subtree of
T. See Fig. 10.5.1.

194
V

V
L

VR

(a)
Positional binary tree T
V
L

(b)
Left subtree T v L
Note : 1) T v L

V
R

(c)
Right subtree T v R

Fig. 10.5.1 Subtree

if it exists is a positional binary tree with root v L

2) T v R if it exists is a positional binary tree with root v R

10.5.2 Methods of tree searching
There are three methods of searching. They are :
1) Preorder Search
2) Inorder Search
3) Postorder Search
We will discuss these three methods of tree searching in
detail.
10.5.2.1 Preorder Search
Preorder search of a tree consists of following three steps :
1) Visit the root
2) Search the left subtree if it exists
3) Search the right subtree if it exists

195

Exercise : Use preorder search and find all subtrees of given tree t in
Fig. 10.5.2(a).
a

k
j

(a) Tree T for preorder search

a

b
k
c

i
5

g
3

4 h

2
1

f
10

(b) Subtrees of T after preorder search

Fig. 10.5.2 a b c g h i d k e j f
Procedure : According to preorder search applied to T, first the root
is visited and a will be printed. Then the left subtree starting at b
will be considered. The boxes represents the subtrees in the order of
their search. The next vertex that will be printed will be b. At b
the left subtree begin at c. Thus, next printed vertex will be c.
After c comes g and h. All the subtrees towards left of vertex
a are searched. Now, we proceed to right subtrees of vertex a.
The order in which search gives the vertex are d then k then e
then j then f. Thus, the preorder search gives us tree T as a b c g
h i d k e j f.

196

10.5.2.1(a) : Polish form or prefix of algebraic expression

Consider the parenthesized expression

p+ q r s d

Fig. 10.5.3(a) shows the labeled positional binary tree

representation of this expression.

s
Fig. 10.5.3(a) Labeled positional binary tree

We can apply preorder search to this tree as shown in Fig.

10.5.3(b).
x

5
p
2

d
7

Fig. 10.5.3(b) Tree x + pq r sd

The preorder search gives the string x + pq r sd. This
is polish form of the given expression ((p + q) (r (s d)). The
boxes shows the order of getting the subtrees.

197

10.5.2.2 Inorder Search

Inorder search of tree consists of following steps.
1) Search left subtree T v L , v L if it exists
2) Visit the root, v.
3) Search the right subtree T v R , v R if it exists.
Exercises : Use inorder search and find all the subtree of tree T
given in Fig. 10.5.2(a).
a

b
i 5

6 k

e
g
1

h
3

10

8
7

Fig. 10.5.4 Inorder search

First, search subtree 1 in Fig. 10.5.2(a). This requires us to
search subtree 2 and this in turn requires us to search subtree 3. As
before, a search of tree simply prints the label of the vertex.
Thus, symbol g is printed first. The root of g is c, so next
printed vertex is c. Right of c is h, so next vertex is h which is
printed. Now, the next subtree begins at b, and right of it is i.
The vertex which will be printed after c will be b and i. The
left subtree of a is complete. Start with the innermost vertex of
right subtree. Subtree in box 6, gives vertex k, then comes root d
and then vertex f. Now vertex f is to right of root e. Finally the
inorder search will give the tree with following order of vertices, g
c h b i a k d j e f.
10.5.2.2(a) Infix notation for algebraic expression
The labeled positional binary tree given in Fig. 10.5.3(a) is
considered for the expression p + q r s d . When inorder
search is applied to this tree we get the string.
p+qr s d
This is infix notation for algebraic expression.

198

10.5.2.3 Postorder Search

Postorder search of tree consists of following steps :
1) Search for left subtree T v L , v L if it exists.
2) Search for right subtree T v R , v R if it exists.
3) Visit the root v.
Exercise : Use postorder search and find all the subtree of tree T
given in Fig. 10.5.2(a).
a

d
b

4
j

3
5

f
8

9
10

Fig. 10.5.5 Postorder Search

When we begin the search we see that subtree 3 and subtree 4
must be searched, before vertex c is printed, if Fig. 10.5.2(b) is
considered for post order search.
Thus the order in which the vertices will be considered in the
postorder search can be given by : g h c i b k j f e d a.
10.5.2.3(a) Reverse polish form or postfix for algebraic
expression
When postorder search is applied to tree of the expression
p + q r s d we get the string pq + rsd x
This is the postfix or Reverse Polish form of the algebraic
expreesion.
Note that in a preorder search the order is root, left and right.
For the inorder search the order is left, root, right while in postorder
search the order is left, right, root.

199

10.6 DIRECTED AND UNDIRECTED TREES :

The trees with directed edges are of great importance in many
applications such as electrical network analysis, game theory, theory
of languages, computer programming and counting problems. One
of the rooted tree with directed edges is called arborescence.
An undirected tree is simply a tree when all the edges are
made bidirectional. We donot use arrows for the edges in undirected
tree. Some examples of undirected trees is shown in Fig. 9.6.1.

(a)

(b)

(c)

Fig. 10.6.1 Undirected trees

Note that if T is an undirected tree then it is connected and
acyclic. The converse also holds. Thus any connected and cyclic
graph is a undirected tree. Some more properties of directed and
undirected trees are discussed in section 10.2.

9.7 EXERCISES :
1) Draw a tree with vertex set V and edge set E. Also find root if
exists.
a) V = {a, b, c, d, e, f}; E = {(a, b), (c, e), (f, a), (f, c), (f, d)}
b) V = {1, 2, 3, 4, 5, 6}; E = {(1, 2), (1, 3), (4, 5), (4, 6)}
2) List all the level 3 vertices and all the leaves of Fig. 10.2.5(a)
and 10.2.5(b).
3) Consider following Fig. 10.7.1.
v0
v

v4

10

v2

11

v 13

12

14

15

Fig. 10.7.1 T(v 0 )

v9

200

a)
b)
c)
d)
e)
f)

Find siblings of v8 .
Find descendants of v8 .
Compute tree T(v 2 ) and T(v3 ) .
What is height of T(v0 ) and T(v3 ) ?.
List all level 4 vertices if exists.
List all leaves.

4) Construct tree for algebraic expression

a) 17 + (6 2) ( x ( y 4))
b) 3 ( x + (6 4 (2 3)))
c) ( x y) (( x 3) (z 4))
5) Make a family tree that shows the descendants of one of your
great grandfather.
6) How many distinct positional 3 trees are there with height 2?
Draw them.
7) Use Prims algorithm to find minimal spanning tree for following
graphs and beginning at vertex given.
B
2.7

a)

2.4

3.4

1.8

2 .9

3.6
3.3
I
J
2.8

2.5

16

15

20

A
12

12

at vertex F

10
15

12

,
4.4

2.1

5.3

b)

2.2

3.2

1.7

4.2

2.1

20
14

10

at vertex G

201
A
4

c)

6
3

5
3

at vertex E.

,
F

8) Use Krushkals algorithm to find minimal spanning tree for the

graphs given in Exercise 7.
9) Refer following tree and answer.
v0
v

v2

v4

v9

10

a) What is height of the tree?

b) List the leaves of T.
c) How many subtrees of T condition v 4 ?
d) List the siblings of v 7 .
10)For the graphs given below perform
i) Preorder search
ii) Inorder search
iii) Post order search and write the result of your search.
x
x
z

s
u
v

(a)

1
z

(b)

202
a
b

c
d

(c)

+
-

r
s

(d)

10.8 LET US SUM UP

In this chapter, we have learnt the structure of trees and its
properties we have also seen different ways of searching trees in a
graph. One can also find the spanning tree using the Prims and
Kruskals algorithm. The trees are very useful models for different
situations in computer science.

10.9 REFERENCES FOR FURTHER READING

1. Discrete Mathematical structures by Kolamn, Busby and Ross.
Pearson education.
2. Introduction to graph theory by Douglas B. West.
3. Discrete Mathematics and its applications by Kenneth. H.Rosen.
McGraw Hill edition.
4. Graph theory by Frank Harary. Narosa Publication.
5. Discrete Mathematics by Norman Priggs. Oxford.

203

11
SEMI-GROUPS AND GROUPS
Unit Structure :
11.0
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9

Objectives
Introduction
Binary Operations
Semi Groups
Products and quotients of semi-group
Groups
Products and quotients of groups
Let us sum up
Unit end exercise
References for further reading

11.0 OBJECTIVES
After going through this chapter students will be able to:
about binary operations
Algebraic structures like semi-groups and groups will be
known
Operations like product and quotient of these algebraic
structure will be known

11.1 INTRODUCTION
Semi-groups and groups are mathematical structures. Semigroups help in the study of finite state machines. While studying
group structure we develop an understanding for coding theory. To
study groups and semi-groups some knowledge of set theory and
number system is required.
In this chapter we are going to discuss following topics.
What are binary operations?
The structure called semi-group, their products and quotients.
The group structure and its product and quotient.

204

11.2 BINARY OPERATION

Binary operation is basic tool to study discrete mathematics.
A collection of objects with operations defined on them and the
properties associated with the operation together gives us a system
which we call mathematical structure or system. An operation that
combines two objects is a binary operation. Binary operation is a
function with certain properties. A set with binary operation is a set
in which an abstract product is defined such that the product of two
elements of the set is again an element of the set.
11.2.1 Definition 1: A binary operation on set G is defined as a
function f : G G G. If a and b G then f(a, b) G.
Remark : A binary operation is a rule which assigns to each ordered
pair of element of G, a unique element of G.
Notation : We use the symbol a * b to denote f (a, b).
11.2.2 Examples of binary operation :
1. Let G = = The set of integers.
Define * : G G G as a b = a + b
Since a + b = G, * is binary operation on .
2. Let G = = set of real numbers.
Define * : as a b = a b
Then * is not binary operation, since it is not defined for every
pair of elements of . For example; 3 and 0 but 3 0 is
not defined.
3. Let G = + = set of positive integers, where * is defined as
a b = a b.
* is not binary operation since it is not defined for every pair of
elements of + . For example; 2 + , 3 + but 2 3 = 1 + .
4. Let G = , be set of integers
Define * : G G G as a b = a + b ab
Then * is binary operation. Note that if G = + = set of positive
integers then * defined above will not be binary operation as 2 + ,
3 + but 2 * 3 = 2 + 3 2.3 = 2 + 3 6 = 1 + .
5. Let M be set of all n n Boolean matrices.

205

A Boolean matrix is n n matrix whose entries are zero or

one. Let A = [aij] and B = [bij] be n n Boolean matrices.
Define A B , the join of A and B, by C = [cij] where
1 if aij = 1 or bij = 1
cij =
0 if aij and bij both are 0

Define A B , the meet of A and B by D = [dij] where

1 if aij and bij are both 1
dij =
0 if aij = 0 or bij = 0

Let M be set of Boolean matrices. Let G = M. Define * on M as

follows : For A, B M; A * B = A B . The * is a binary operation.
If * is defined as A * B = A B then, again * is binary operation.
11.2.3 Properties of binary operation :
1. Definition 2: A binary operation on a set G is said to be closed if
a * b G for all elements a and b in G. We say * satisfies closure
property.
Note : Whenever * is binary operation, it always hold closure
property and we say G is closed with respect to *.
2. Definition 3: A binary operation on set G is said to be
commutative if a * b = b * a for all a, b G. We say * satisfies
commutative property.
3. Definition 4: A binary operation * on a set G is said to be
associative if a * (b * c) = (a * b) * c for all a, b, c G. We say *
satisfies associative property.
4. Definition 5: A binary operation * on a set G is said to be
idempotent if a * a = a for all a G . We say * satisfies idempotent
property.
To summaries properties of binary operation we have
following table where * is binary operation on a set G and * satisfies
properties for a, b, c G.
1.
2.
3.
4.

a*bG
a*a=a
a*b=b*a
a * (b * c) = (a * b) * c

Closure Property
Idempotent Property
Commutative Property
Associative Property
Table No. 11.1

206

Examples based on definition 2 to 5.

1. Let G = and a * b = a + b, a, b , Then * satisfies closure
property as a * b = a + b .
* satisfies commutative property because a * b = a + b while b
* a = b + a and a * b = a + b = b + a = b * a.
Also, (a * b) * c = (a + b) * c
= (a + b) + c
= a + (b + c) isassociativein
= a + (b * c)
= a * (b * c)

* holds associative property. Now, 2 * 2 = 2 + 2 = 4 2, hence *

does not satisfy idempotent property.
2. Let L be a lattice. Define a * b = a b (greatest lower bound of
a and b) Then * satisfies all four properties.
* holds closure property
a * b = a b L,
* holds idempotent property
a * a = a a = a,
a * b = a b = b a = b * a * holds commutative property
(a * b) * c = (a b) c
= a (b c)
= a * (b * c), * holds associative property
3. Let G and * be defined as, a*bab2b
Then a * b , hence * holds closure property.
a * a = a.a + 2a = a (a + 2) a, hence * does not hold idempotent
property.
a * b = ab + 2b and
b * a = ba + 2a
Since a * b b * a, * is not commutative on .
(a * b) * c
= (ab + 2b) * c
= (ab + 2b) . c + 2c
= abc + 2bc + 2c

and

a * (b * c)
= a * (bc + 2c)
= a (bc + 2c) + 2 (bc + 2c)

207

= abc + 2ac + 2bc + 4c

Thus, (a * b) * c a * (b * c) and hence * is not associative on .
4. Consider the set G = {a, b, c, d} with binary operation * defined
by following table.
*

a
b
c
d

a
a
d
c
d

b
c
a
d
b

c
b
b
a
a

d
d
c
a
c

i) c * d = a and d * c = a
Thus, c * d = d * c
ii) b * d = c and d * b = b
Thus, b * d d * b
iii) a * (b * c)
=a*b
=c
(a * b) * c = c * c = a
a * (b * c) (a * b) * c
From (ii), * is not commutative.
and from (iii), * is not associative.
Check your progress
1. Determine whether * is binary operation. If it is determine
whether * is closed, idempotent, commutative and associate on
given set.
a) On where a * b = 2a + b
b) On + where a * b = a / b
c) On where a * b = a b
d) On lattice L where a * b = a b
(least upper bound of a and b)
2. Consider binary operation * defined on set G = {a, b, c} given by
following table.
*

a
b
c
a)
b)
c)

a
b
c
b
c
b
a
b
c
c
a
b
Is * commutative?
Compute a * (b * c) and (a * b) * c
Is * associative?

208

11.3 SEMI-GROUPS
Definition 6 : Algebraic Structure
A nonempty set G with one or more binary operations is called
an algebraic structure.
If * is binary operation on G then (G, *) is an algebraic structure.
Examples of algebraic structure
1. The set : set of natural numbers is algebraic structure with
respect to binary operation +. Thus we denote ( , +) is an
algebraic structure.
2. ( , +) : set of integers with binary operation + is an algebraic
structure.
3. ( , + , ) : set of real numbers with binary operations + and ,
is an algebraic structure.
Definition 7: Semi-group :
An algebraic structure (G, *) is called a semi-group if the binary
operation * is associative in G. Thus, if a, b, c G, then (a * b) * c =
a * (b * c).
Definition 8: Commutative Semi-group :
The semi-group (G, *) is said to be commutative if * is commutative.
Examples of Semi-group :
1. ( , +) : Set of natural numbers with respect to binary operation
+ is semi group as + satisfies associative property i.e. a, b, c
, (a + b) + c = a + (b + c)
2. ( , +) : set of integers with binary operation + is commutative
semi-group because + is associative and commutative in .
Definition 9: Identity element:
An element e in a semi group (G, *) is called the identity element if
e * a = a * e = a a G. (read as for all)
Note that identity element is unique. Otherwise if it is not unique
then there exist another identity element i such that i * a = a * i = a.
Thus if a = e then; i * e = e * i = e
Also, if a = i then; e * i = i * e = i

(1)
(2)

209

From (1) and (2) we get e = i. Thus, identity element if it exists is

unique.
Definition 10: Monoid
A monoid is a semigroup (G, *) that has an identity element.
Examples of Monoid
1. ( , +) : Set of integers with binary operation + is monoid. Here;
0 is identity element as 0 + a = a + 0 = a a .
2. Let S be fixed non empty set and let Ss be set of all functions
f : S S. If f and g are elements of Ss , define f * g = f g, the
composite function. Then ( Ss , *) is a semigroup which is not
commutative and is a monoid since Ss has identity element 1s , i.e.
f Ss ; 1s * f = f *1s = f

Definition 11: Sub semi-group

Let (G, *) be a semi-group. Let H be subset of G. If H is
closed under binary operation * then (H, *) is called sub-semi-group
of (G, *)
Definition 12: Sub monoid
Let (G, *) be a monoid with identity element e. If H be nonempty
subset of G. If H is closed under binary operation * and e H, then
(H, *) is called submonoid.
Note : 1) Subsemigroup of a semigroup is itself a semigroup.
2) Submonoid of a monoid is itself a monoid.
Examples of submonoid
1. If (G, *) is a semigroup, then (G, *) is subsemigroup of (G, *).
Similarly if (G, *) is a monoid then (G, *) is submonoid of (G, *). If
T = {e} then (T, *) is also a submonoid of monoid (G,*).
2. Let H be set of all even integers then (H, X) is a sub semigroup
of ( , X) where x is binary operation multiplication. But (H, X) is
not a submonoid of ( , X) because 1 is identity element of which
does not belong to H.
Group Theory
A group is formally defined as below. We denote the binary
operation as 0 or * until or otherwise specified.

210

Definition: Let G be a non-empty set and 0 be a binary operation on

G. We say that (G, 0) is a group if the following four properties are
satisfied.
1. G. is closed with respect to * i.e .,for all a , b in G the element a
* b is a uniquely defined element of G
2. G. is associative with respect to *
i.e., for all a , b , c in G a * (b*c) = (a*b) *c
3. Identity element exists in G for *
i.e., if there exists e such that a * e = e * a = a a in G.
4. Inverse exists for each element in G with respect to 0
i.e., for each a in G there exists an element a -1 in G such that
a*a1a1 * ae (where e is identity element of G)
Example 1: Set of all non-zero rational numbers from a group under
ordinary multiplication.
Solution: Let Q* is the set of all non-zero rational numbers.
Closure law: Let a , b Q*
a . b also belongs to Q* (Product of two rational numbers is a ational
number)
Q* is closed with respect to multiplication.
Q* satisfies fist condition of a group
u
v
w
Associative law: Let a, b, c Q * let a , b and c
x
y
3
uv w uv w u vw w
u v uv
Consider a b ;a.b.c

xy z xy z
x y xy
x yz
( (uv)w = u(vw) and (xy)z = x(yz) where u,v,w,x,y,z Z and
satisfies associative law)
u v w

x y z

a b.c
From the above example it is clear that (a . b) . c = a .(b.c) and it si
true a , b , cQ*
Q* is associative with respect to multiplication.
Existence of Identity: since 1 is a rationa1 number 1 Q*
Let a Q*
We have a.1 = a = 1.a
So 1 is the identity element of Q* which exists in Q*

211

Existence of Inverse: To prove existence of inverse, let a

element Q* There exists

p
be an
q

q
p q
Q * such that 1
p
q p

q
p
is the inverse of and it is true for all a Q*
p
q
Hence inverse exists for such element of Q* and inverse of a
denoted by 1/a or a-1.
That shows

Q* satisfies all four properties of a group, Q* is a group under

multiplication (Q*, .) is a group
Properties of a Group :
Abelian Group: In addition to the above mentioned four properties
of a group if it also satisfies another property called commutative
property , i.e.,a*b = b*a a , b G
The group is calles either Abelian group or commutative group. A
group which is not abelian is calles as non-abelian group
Example 3: C = { a + ib / a,b R} C is an abelian group with two
addition.
Addition is defined on C as (a1 + i b1) + (a2 + i b2) = (a1 + a2)
+ (b1 + b2)
Commutative Property
Let x , y C where x = a1 + i b1 and y = a2 + i b2 , a1 , a2 , b1 , b2R
X + y = (a1 + i b1) + (a2 + i b2)
= (a1 + a2) + i(b1 + b2)
Since addition two real number satisfies the commutative law.
a1 + a2 = a2+a1 and b1+b2=b2+b1
= (a2 + a1) + i(b2 + b1)
= (a2 + i b2) + (a1 + i b1)
=y+z
C satisfies the commutative law with respect to addition.
(C , +) is an abelian group or (C , +) is a commutative group.
Set of complex number also forms an abelian group with respect to
multiplication.
(Leftas an exercise)

212

ab ba
b * Aabba
5
5
Hance Q+ is an abelian group with respect to *.
Commutative Law : a, bQa * b

Example 4: G = {1, -1} is an abelian group under multiplication.

.
1
-1

1
1
-1

-1
-1
1

From the above table, it is clear that (G , .) satisfies both closure,

associative property ,and abelian property with 1 being identity and 1 is its own inverse
Addition modulo m
We shall now define a new type of addition called addition
modulo and is denoted by a+m b where a and b are integers and m
is a fixed positive integers.
By definition , a+m b = r 0 r < m where r is the least non-negative
remainder when a + b is devided by m and we read itas a addition
modulo m b.
Example 5: If a = 7; b = 8 then add 7 and 8 gives 15 divide by 2
the remainder is 1.
7+2 8=1
If a = 5 and b = 6 add 5 and 6 gives 11 divide by 3 we get the
remainder is 2.
5+3 6=2
Note: If a and b are two integers such that a-b divisible by fixed
positive integer m we write a = b (mod m) and we read it as a is
congruent to b modulo m
Note: It can be easily seen that a+m b = b+m c (take any example and
try it in your own)
Example 6: prove that the set G = { 0, 1, 2, 3, 4, 5} is a finite
abelian group of group of order 6 with respect to addition
modulo 6.

213

From the composition table as shown below

1
2
3
6 0
0
06 00 06 11 06 22 06 33
1 6 0 1
1 6 1 2
1 6 2 3
1 6 3 4
1
2
26 02 26 13 26 2 4 26 35
3
36 03 36 14 36 25 36 30
4
46 04 46 15 46 20 46 3 1
5
56 05 56 10 56 21 56 32

06 44

06 55

1 6 4 5

1 6 5 0

36 41
46 42

36 52
46 53

26 40

56 43

26 51

56 54

From the above table we see that all entries in the composite table
are the element of G.
That shows G is closed under addition modulo 6.(+6.)
To prove G is associative, let a = 2 b = 4 c=1
Consider 2+6 (4+6 1) = 2+65 = 1
(2+6 4) +6 1= 0+61 = 1
2+6 (4+6 1) = (2+6 4) +6 1 and it is true a , b , c G
G is associative under addition modulo 6 (+6.)
Existence of identity : let a G a +6 0 = 0+6 a = a a G.
0 is the identity element in G.
Existence of inverse: from the above table,
0+60=0
1+65=0
2+64=0
3+63=0
4+62=0
5+61=0
Inverse of 0 is 0 , inverse of 1 is 5 , inverse of 2 is 4 , inverse
of 3 is 3 , inverse of 4 is 2 , inverse of 5 is 1
inverse exists for
each element of G and belongs to G
G is a group with respect to the binary operation +6.
Commutative Law : a +6 b = b +6 a a, b G.
If a = 2 b = 4 2+6 4 = 0 = 4+6 2.
(G, +6) is an abelian group)
Note: The set of first m non-negative forms an abelian hroup with
respect to addition modulo m

214

Finite and identity Group: If the set G contains a finite number of

element then the group then the group (G, *) is called a finite group.
Otherwise the group (G, *) is called as Infinite group
Order of a group: Another natural characteristic of a group G is the
number of element it contains. We call it as order of a group and is
denoted by O(G).
Example 10: Let G = {1, -1} is a group
Then O(G) = 2.
If G is a group containing the set of all integers or set of all natural
numbers, then O(G) is infinite.
Order of an element of a group: If G is a group and a G. The
order of a is the least positive integer m such that am = e.
So, to find the order of a group element compute a,a2,a3,.. until you
reach the identity for the first time. See the following example.
In the group { 1 , -1 , I , -i} 1 is identity element i1 =I , i2 = -1 , i3 = -I
, i4 = 1 , i5 =I , i6 = -1 , i7 = -I , i8 = 1. Identity appeared twice at i4 and
i8 , but o (i) = 4 ( it is the least)
If such integer does not exists we say that the order of a is infinity.
We use the notation O (a) for the order of a.
Co-prime
Two number are said to be co prime if they do not have any common
factory except 1. If a are co primes then there exists two integers x
, y such that xa + by = 1.
Example:-7.
On a group G, O (a) = 18, State that the orders of a6, a15, a-7.
Solution:- Oan

n
n, k

18
18

18,6 6

O a k
1) O (a) = 18

O a 6

215

2) O (a) = 18

18
18

18,15 2

18
18,7

O a15
3) O (a) = 18

O a7

O a7 O a 7 O (a
Now O a7

-7

) = O (a7) (the order of element of group is

same as its inverse )

18
O a7
18 .
18, 7

Example 11: Find the order of such element of the group G = {1, 2,
4, 7, 8, 11, 13, 14} the composition being multiplication modulo 15.
Solution: Identity element of G = 1:O (1) =1
To Find the order of 2,2x152x152 = 4x152=8
2x152x152 = 4x152=8
2x152x152x152 = 8x152=1(identity)
Hence O(2) = 4
To Find the order of 4,4x154 =1(identity)
Hence O(4) = 2
To Find the order of 7,7x157 = 4
7x157x157 = 13
7x157x157x157 = 13x157 = 1(identity)
Similarly, we can compute the order of 8,11,13,14.
Sub Group :
Sub Group: In general we are not interested in a subset of a group
G. but certain subset of elements in a group is itself a group. This
situation arises so often that we introduce a special name to describe
it, called sub group. See the following definition for a subgroup.
Definition: A non-empty subset H of a group G is said to be a
subgroup of G if H itself is a group, with respect to the same binary
operation defined on G.
Every subgroup of G is a complex of G every complex is not always
a subgroup.

216

Example 12: Q*under multiplication is a proper subgroup of R*

under multiplication.
Example 13: Additive group of even integer is a subgroup of the
additive group of all integers.
Two-Step Subgroup Test:
Theorem: A non-empty subset of H of a group G is a subgroup of G
if and only if
i. a,b H implies ab H
ii. a H implies that a-1 H
Theorem: A non-empty subset of H of a group G is a subgroup of G
if and only if a , b H
ab-1 H.
One-Step Subgroup Test
Example 14 : See the example to under stand one-step subgroup
Test.
G be the group of non-zero complex numbers under
multiplication.
H = {a+ib/a2+b2 = 1,aR bR} is a sub group of G.
Let x , y H where x = a + ib and y = c + id
Inverse of y

1
C id
2
C id C d 2

we have xy-1 = (a + ib)

real part of xy1

C id

C d
2

= ac+bd +i(bc-ad)/c2 + d2

ac id

C2 d 2

imaginary part of xy1

bc ad

C2 d 2

ac bd bc ad
consider 2

C d 2 C 2 d 2
2
2

a 2c 2 b 2 d 2 2abcd b2 c 2 a 2 d 2 2abcd

c 2 d 2
a 2 b 2 c 2 d 2 1.1

1
2
2
1
c

Xy-1 H , hence H is a subgroup of G.

217

Example 15: Let Z be the group of all integers

Let H1 = {..-6,-4,-2, 0, 2, 4, 6}
H2 = {..-12,-9,-6,-3, 0, 3, 6, 9, 12} be two subgroups of Z.
H1 H2 = { -12, -9, -6, -4, -3, -2, 0, 2, 3, 4, 6, 9,}
Since -2 H1 H2 , -3 H1 H2 but -2 + -3 H1 H2
H1 H2 , is not closed under +.
H1 H2 , is not subgroup of (Z, + ).
Example 16 : G = {1, -1} is a group
H = {-1} is a subset of G.
H-1 = {-1}
H = H-1 but H is not a group since identity does not exists.
H is not a subgroup of G.
Check your progress
1. Show that AB is a sub-group of G if and only if AB = BA.
2. When does the semi-group form a group?
3. Prove that the set {0, 1, 2, 3, 4, 5} is a finite abelian group of
order 6 with respect to addition modulo 6.
4. Show that the set G {a b 2 : a, b Q} is a group with respect to
addition.

11.7 LET US SUM UP

In this chapter we have learnt the details of algebraic
structures semi-group and groups. The examples of semi-group and
group are varying because of the properties related to the structure.
The study of semi-groups and groups will make the study of finite
state machines and coding theory simpler.

11.8 UNIT END EXERCISE

Q.1

Show that the set of all positive rational numbers forms an

abelian group under the composition defined by a b

Q.2
Q.3
Q.4

ab
.
2

Show that set IN of all natural numbers is not a group with

respect addition.
Find the order of the elements of the group ( z4 , 4) .
Find the order of the elements of the group ({1, w, w2}, . ).
Where w isa cube root of 1.

218

Q.5

Prove that the fourth roots of unity form an abelian group

under multiplication.
Q.6 Prove that the set {0, 1, 2, 3, 4} is a finite Abelian group of
order 5 under addition modulo 5 as composition.
Q.7 Check whether (Z, -) is semi-group or not. Where - denotes
integer subtraction.
Q.8 Check whether (Z, +) and (Z, .) are monoids or not.
cos sin
Q.9 Show that the set of matrices A
, where
sin cos
IR , forms a group under matrix multiplication.
Q.10 Show that the integer multiples of 5 form a sub-group of the
additive group of integers.

11.9 REFERENCES FOR FURTHER READING

1. University Algebra by N.S. Gopalkrishnan.
2. Contemptary Algebra by Gallian.
3. Discrete mathematics by Kenneth and Rosen.
4. Discrete mathematical structures by Kolman, Busby and Ross.
5. Modern Algebra by Vasishtha.

219

12
NORMAL SUBGROUP
Unit Structure
12.0 Objectives
12.1 Introduction
12.2 Product and quotient of algebraic structures
12.3 Homomorphism
12.4 Isomorphism
12.5 Automorphism
12.6 Cyclic groups
12.7
Normal Subgroup
12.8
Codes and group code
12.9 Let us sum up
12.10 Unit end exercise
12.11 References for further reading

12.0 OBJECTIVES :
After going through this chapter students will be able to know:

Operations like product and quotient of these algebraic

structures.

Isomorphism, Homomorphism and Automorphism group.

Generators of Cyclic group.
Normal sub-group.
Coding and Encoding of group.

12.1 INTRODUCTION :
After having all the basic property of group and sub-group,
we now begin our journey with more detail about group study. In
this we are going to discuss about product of group and quotients
group, isomorphic group, homomorphic group, automorphic group.
In group theory cyclic group are the simplest group also it is very
interesting. In previous chapter we learn about sub-group, now
here we discuss about cosets and normal sub-group.

220

12.2

PRODUCT AND QUOTIENTS OF GROUPS :

12.6.1 Definition : If G1 and G 2 are groups then the product of G1

and G 2 denoted as G1 G 2 is a group with binary operation
defined by a1, b1 a 2 , b 2 = a1a 2 , b1b 2
Examples 1: Let G1 = G 2 = 2 = 0, 1 , 0 is notation for [0] find
G1 G 2 .

Solution: G = G1 G 2 = 2 2
=

0, 0 , 0, 1 , 1, 0 , 1, 1

Composition table for 2 2 is

x

0, 0
1, 0

0, 1
1, 1

0, 0 1, 0 0,
0, 0 1, 0 0,
1, 0 0, 0 1,
0, 1 1, 1 0,
1, 1 0, 1 1,

1, 1
1 1, 1
1 0, 1
1

1, 0
0 0, 0

Note that in ( 2 , +), 1 + 1 = 2 = 0 . G is group of order four.

12.3 HOMOMORPHISM
Group of Homomorphism: Till now we have seen the notion of a
group and various type of group. Now we see the relation between
two groups by introducing Homomorphism. A relation between
groups G and G, is generally exhibited in terms of a structure
relating map from G to G1.
Let G and G1, be two groups. We are interested in a map that
relates the group structure of G to the group structure of G1, and this
map often gives us information about the structure of G1 from
known structural properties of G, or information about the structure
of G from known structural properties of G1
We known that the group structure is determined by its binary
operation. We now define such a structure relating map for groups,
and then point out how the binary operations of G and G1 are related
by such a map.

221

Definition: Let G and G1 be groups. A map f : G G1 is said to be

Homomorphism
If f(ab) = f(a) f(b) for all a , b G.
Note: If the operation in G is denoted by * and the operation in G1
is The above condition for Homomorphism means the following.
f(aob) = f(a) . f(b).
12.3.1 Properties of Homomorphism
Let G and G1 be two groups. e and e1 be the identity element of G
and G1 respectively. If f is Homomorphism from G to G1 then
f(e) = e1
Range of Homomorphism: G and G1 are two groups and f is
homomorphism from G to G1. The set of all f images of G in G1 is
called range of homomorphism.
It can be written as f (G) = { f (a) / a G}
12.3.2 Types of Homomorphism
Onto Homomorphism: Let G and G1 be two groups and f is a
mapping from G onto G1.
If f (ab) = f (a) f (b) a, b G then f is said to be a Homomorphism
from G onto G1.
In some books it is referred as epimorphism.
Endomorphism: A homomorphism of a group into itself is called
an endomorphism.
Monomorphism: If the homomorphism is one-one it is called
monomorphism.
Example 2: Let G be the additive of integers and G1 be the
multiplicative group. Show that f :G G1
a function defined as f (m) = em is a homomorphism
Solution: Let m , n G ; f (m) = em G1 and f (n) = en G1
m+n G (G is additive group)
f (m+n) = em+n = em en = f (m) f (n)
f is homomorphism from G to G1.

12.4 ISOMORPHISM
Isomorphism: A function f from G to G1. Is said to be isomorphism,
if
1. f :G G1 is one-one
2 . f: G G1 is onto

222

3. f: G G1 is homomorphism.
Says distinct element in G have distinct f - images in G1
Says X G1aG such that f a x .
Says image of the product is same as product of images.
Note: in the above definition, we have denoted the operation as
multiplication. We can use different symbols to denote the
compositions.
Note:There may exists more than one isomorphism from G onto G1.
Example 3: Let G be the multiplication group of all positive real
numbers, and G1 be the additive group of all real numbers. The
mapping defined by f : G G1 such that f (x) = log x .is
isomorphism from G to G1
Solution: : G G1 = log x
To prove

is one one

Let x1, x2 and

(x1) =

(x2)

Log x1 = log x2
elogx1 = elogx2
x1 = x2

is one one from G to G1

To prove
that ey G

is on-to

For any real number yG1 ey is a positive real number such

(ey) = log ey = y G1
Each element of G1 is the
i.e.

is on-to.

To prove

as homomorphism.

Consider x,y G where

Then

- image of some element in G.

(x) = log x :

(xy) = log (xy)

= log(x) + log(y)
=

(x) +

(y)

is homomorphism from G to G1
`

is isomomorphism from G to G1

(y) = log y

223

Example 4: There exists isomomorphism from an additive group of

integers.
G = {.,-3, -2, -1, 0, 1, 2, 3} to another additive group
G1 = {.,-3m, -2m, -m, m, 1m, 2m, 3m} where m is any fixed
integers not equal to zero.
Solution: Define mapping : G G1 such that (x) = mx
To prove is one- one : let x1, x2 G and (x1) = (x2)
mx1= mx2 (from the definition of )
x1 = x2
is one one from G to G1
To prove is onto: for any element y G1
y/m G such that
(y/m) = m(y/m) = y G1
each element of G1 is the - image of some element in G.
i.e., os on-to.
To prove as n=homomorphism.
Consider x,y G where
(x) = mx :
(y) = my
(x + y) = m(x+y)
= mx + my
= (x) + (y)
is homomorphism
Is isomorphism from G to G1

12.5 AUTOMORPHISM OF A GROUP

Definition: If f : G G is an isomorphism from a group G to itself,
then f is called an automorphism of G.
Example 5: If G is an additive group of complex number, show that
f : G G such that f(Z) = pZ where p is a non-zero complex
number, an automorphism of G.
Solution: G is an automorphism if f : G G is an isomorphism.
To prove f is one-one.
Let Z1, Z2 G and f(Z1) = f(Z2)
PZ1 = pZ2
Z1=Z2
f is one-one.
To prove f is onto.
For any element Z G there exists Z/p G such that f(Z/p) =
p.(Z/p) = Z.
Each element of G is the f- image of some element in G.
There fore f is onto.
To prove f is homomorphism.

224

Consider f(Z1 + Z2) = p(Z1 + Z2) = pZ1 + pZ2 = f(Z1) + f(Z2)

f is an homomorphism.
There fore f is isomorphism from G to G.
Hence f : G G is an automorphism.

12.6 CYCLIC GROUP

Cyclic group: In the group theory, cyclic group are the simplest are
the simplest among all group all groups. Because of this cyclic
groups possess interesting properties. With the help of cyclic group
we can find answer for some of the difficult questions in group
theory. Now let us see what do we mean by a cyclic group?
The formal definition of a cyclic group is given below.
Definition: A group G is called cyclic if for some a G, every
element x G is of the from such that an
Where n is some integer. The element a is called a generator of G.
A cyclic group G generated by a can be represented as G = <a>
If G is a group with respect to the binary operation addition, cyclic
group is defined as G = { na / n Z}
Example 6: G = {1, -1} is a cyclic group generated by -1 (
1)2, -1 = (-1)1)

1 = (-

12.6.1 Cyclic Subgroup: A subgroup H of a group G is called a

cyclic subgroup if H is a cyclic group.
Note: If a is a generator of a cyclic group G then a-1 is also a
generator
Let G be a cyclic group generated by a.
Then for every x G there exists an integer, such that x = am
= (a-1)-m
Every x can be expressed as integral power of a-1
i.e.,a-1 is also a generator of G.

12.7 COSETS
Definition: If G is a Group and H is a subgroup of G. let a G
Then
Ha = {ha : hH} is called right coset of H in G generated by a.

225

And the set aH = {ah: h H} is called left coset of H in G generated

of a.
Example 7: Let G = { a, b, c, d, e, f} is a group. And H = { b, c, e}
be the subgroup of G.
Solution: a G : Ha = { ba, ca , ea}
d G : Hd = { bd, cd , ed} are some right coset of H in G.
c G : Hc = { bc, cc , ec}
The set aH = { ah : h H} is called left coset of H in G
generated by a.
a G : aH = { ab, ac , ae}
d G : Hd = { db, dc , de} are some right coset of H in G.
c G : Hc = { cb, cc , ce}
Note: If G is an abelian group then aH = Ha
Example 8: Let G be the additive group of integers, and H is a
subset of group of G where element of H are obtained by
multiplying each element of G by 2.
Solution:
Clearly (H, +) is a subgroup of (G, +).
Now G = {..-3,-2,-1, 0, 1, 2, 3}
H = {..-6,-4,-2, 0, 2, 4, 6}
1 G and 1 + H = {..-5,-3,-1, 1, 3, 5, 7}
Example 9: Let G= <a> a cyclic group of order 15. List all the cosets
of <a5 > in G.
Solution: Let G = <a> a cyclic group of order 15.
i.e. G = {e, a, a1, a2, .., a14}
H is a subgroup of G.
H = {e, a5, a10}
The left cosets of <a5> are
aH = {a, a6 , a11},

a2H = {a2, a7, a12},

a3 H = {a3, a8,a13},

a4 H = {a4,a9,a14} .
Remark: A coset may not essentially a subgroup
Remark: If e is the identity in G, it is also identity in H. Then eH =
{eh/h H} = H

226

12.7.1 Normal Subgroup

If G is a group and H is a subgroup of G, it is not always true that
aH = Ha for all a in G. If it happens i.e.,if Ha = aH
a in G we call
H as normal subgroup. And it is denoted by H G.
Definition: If G is a group and H is a subgroup of G.If Hx = xH
xG then H is called a normal subgroup.
Note: Every normal subgroup but every subgroup need not be a
normal subgroup.
Importer and proper normal subgroups: G is a group then G , {e} are
subgroup of G and they are also normal subgroups of G. These two
subgroups are called trivial or importer subgroups of G
The normal subgroup of G other than these two subgroups are called
proper normal subgroups of G.
For example H = {1,-1} is a normal subgroup of multiplicative
group of none zero real numbers.
Example :- Show that every subgroup of an Abelian group is
normal.
Solution :- Let G be an abelian and H a subgroup of G. Let x be any
element of G and H any element of H.
So, xhx-1 = xx-1
=h [G is Abelian x-1h = hx-1]
=eh
h H.
Hence x G,
hH
xhx-1 H
So H is normal in G.
Example :- Given that H = { I, (12)(13) (34)} is a subgroup of A4.
Show that (243)H = (142)H and (132)H = (234)H but (234)(132)H
(142)(234)H. Is H a normal subgroup of Ah? Justify your answer.
Solution :-

H = { I, (12)(34)} is a subgroup of Ah.

A4 = { I, (12)(34)(13)(24)(23)(14)(123)(132)(142)(124)}
To show that
(243)H = (142)H
(243)(12)(34) = (142)
(243)H = {(243)(142)}
(142)H = i.e. (142)I and (142)(12)(34)

227

(142)H = {(142)(243)}
.. (243)H = (142)H
To show that (132)H = (234)H
(132)H = (132)I
(132)(12)(34) = {(132), (234)}
(234I = (234)
(234)(12)(34) = (132)
(234)H = {(234)(132)}
. (132)H = (234)H
To show that (234)(134)H (142)(234)H
(234)(132)H = (234)(132)
(234)(132)(12)(34) = I
(142)(243)(12)(34) = (124)
Thus (243)(132)H (142)(243)H
GH = HG
H is normal subgroup A4.

12.8 CODE AND GROUP CODE

Word : A sequence of 0s and 1s is called a word.
e.g. 1101, 101, 00100 are words.
Code : A collection of words used to represent different messages is
called code.
Codeword : A word in a code is called codeword.
Block of code : A code consisting of words having same length is
called block of code.
Let B = {0, 1} then B B 00, 01,10,11 B2 i.e. B2 contains
words of length 2, and it contains 4 elements or codes. Number of
elements in the set is called cardinality of the set and it is denoted by
two vertical bar.

Number of elements in set B2 | B2 | 4 .

| B2 || B B || B | | B |2 2 22
| B2 | 22 4

Also |B| = 2.

Similarly
B3 = B B B = {000, 001, 010, 011, 100, 101, 110, 111},
| B3 | = | B B B | = | B | | B | | B | = 2 2 2 = 23 = 8

228

The set Bm is collection of codes of length m and it contains

2m

codes.

| Bm |2m
Weight : Let x B n then weight of x is number of 1s in x and it is
denoted by w(x).
e.g. i) x = 1101
B 4 w(x) = 3
ii) x = 110010 B6 w(x) = 3
iii) x = 11
B 2 w(x) = 2
iv) x = 0000
B 4 w(x) = 0
x y : (Read as x ring sum y). Let x, y B n , then x y is a

sequence of length n that has 1s in those position x and y differ and

0s in those positions x and y are the same.
i.e. The operation + is defined as
0+0=0
0+1=1
1+1=0
1+0=1
e.g.
i) x, y B3 , x = 101, y = 110
x =
1 0 1
y =
1 1 0
xy = 0 1 1

x y = 011 and w x y = 2
ii) x, y B6 , x = 110100, y = 111111
x =
1 1 0 1 0 0
y =
1 1 1 1 1 1
xy = 0 0 1 0 1 1
x y = 001011 and w x y = 3
iii) x, y B7 , x = 1010001, y = 0001010
x y = 1011011 and w x y = 5
Distance : The distance between x and y is the weight of x y . i.e.
w x y , it is denoted by d(x, y). The distance between two words
is exactly the number of positions at which they differ.
d(x, y) = w x y .

229

It is also called Hamming distance. Minimum distance: Let

x, y Bn then minimum distance = min. {d(x, y) : x, y Bn }.
Let x1, x 2 ,..., x n are the code words, let any x i , i = 1, 2, n
is a transmitted word and y be the corresponding received word.
Then y x k if d x k , y is the minimum distance for k = 1, 2, n.
This criteria is known as the minimum distance criterion.
Encoding function : Let m < n (m, n N, N is set of natural
numbers) then an one to one function e : B m B n is called an (m, n)
encoding function. i.e. for x B m we have y Bn such that
e(x) = y.
Detection of errors : Let e : B m B n (m < n) is an encoding
function then if minimum distance of e is (k + 1) then it can detect k
or less than k errors.
Correction of errors : Let e : B m B n (m < n) is an encoding
function then if minimum distance of e is (2k + 1) then it can correct
k or less than k errors.
Example 1 : Let e is (2, 4) encoding function defined as
e(00) = 0000
e(01) = 1011
e(11) = 1100

e(10) = 0110

i) Find minimum distance,

ii) How many errors can e detect,
iii) How many errors can e correct.
Solution :
Let x 0 = 0000, x1 = 1011, x 2 = 0110, x 3 = 1100
i) w x 0 x1 w x1 3
w x0 x2 w x2 2

w x0 x3 w x3 2
w x1 x 2 w 1101 3
w x1 x 3 w 0111 3
w x 2 x 3 w 1010 2

Minimum distance of e = 2.

230

Note that minimum distance is not unique. There are three pairs
having distance 2.
ii) k + 1 = 2 k = 1,
e can detect 1 or less than 1 i.e. 0 errors.
iii) 2k + 1 = 2 k =

e can correct

1
2

1
1
or less than
errors, i.e. e can correct 0
2
2

errors.
Example 2 : Let e is (3, 8) encoding function defined as
e(000) = 00000000
e(011) = 01110001
e(010) = 10011100

e(110) = 11110000

e(001) = 01110010

e(101) = 10110000

e(100) = 01100101
e(111) = 00001111
i) Find minimum distance.
ii) How many errors can e detect?
iii) How many errors can e correct?
Solution :
Let x 0 = 00000000, x1 = 10011100, x 2 = 01110010, x 3 =
01100101, x 4 = 01110001, x 5 = 11110000, x 6 = 10110000,
x 7 = 00001111.
i) w x 0 x1 w x1 4 ,
w x0 x2 w x2 4 ,
w x0 x3 w x3 4 ,

w x0 x5 w x5 4 ,
w x0 x7 w x7 4

w x0 x4 w x4 4 ,

w x0 x6 w x6 3 ,

Similarly, w x1 x 2 w 11101110 6 ,

w x1 x 3 6 , w x1 x 4 6 , w x1 x 5 4 , w x1 x 6 3 ,

w x1 x 7 4 , w x 2 x 3 4 , w x 2 x 4 2 , w x 2 x 5 2 ,
w x 2 x6 3 , w x 2 x 7 6 , w x3 x 4 2 , w x3 x5 4 ,

w x3 x 6 5 , w x3 x7 4 , w x 4 x5 2 , w x 4 x6 3 ,

w x 4 x 7 6 , w x5 x6 1 , w x5 x 7 8 , w x6 x 7 7

The minimum distance of e = 1.

ii) k + 1 = 1 k = 0
e can detect 0 or less than 0 errors i.e. 0 errors.
iii) 2k + 1 = 1 k = 0
e can correct 0 or less than 0 errors. i.e. 0 errors.

231

Example 3 : Compute
1
0

1
1
0
0

0
1

1
1

0
0

0
1

0
0
0
1

0
1
1

Solution :
1 1
0 1

1 0

0 1

1 0
1 0
00
0 1

0 0
0

1
11

1
0 1

0 0
1

1
1
0
1

0
0
1

Same digit sum = 0, opposite digit sum = 1

Example 4 : Let B = {0, 1} and + is defined on B as follows.
+
0
1
0

Then show that (B, +) is a group.

Solution :
Addition is associative. Here B is set of bits and the operation of on
B is +. B with operation + is associative.
Also 0 + 1 = 1 and 0 + 0 = 0
0 B is an identity element. Here inverse of each element is
itself. Since 0 + 0 = 0. 0 1 0
11 1
and 1 + 1 = 0
Inverse of each element exists.
(B, +) is a group.
Three Cartesian product of groups is again a group.
Bn B B B...ntimes... B with + operation defined as
x1,x 2 ,...,x n y1,y 2 ,...,y n x1 y1,x 2 y 2 ,...,x n y n is also a
group. Here identity element is (0, 0, 0) Bn and every element
is its own inverse.

Bn , is a group.

Let A Bn such that A, is a group then

A is subgroup of Bn . Now we will see the encoding which uses this

property of Bn .

232

Check Your Progress :

1. Let e : B 2 B6 is an (2.6) encoding function defined as
e(00) = 000000 e(01) = 011101
e(11) = 111111 e(10) = 001110
i) Find minimum distance.
ii) How many errors can e detect?
iii) How many errors can e correct?
2. Let e is (2, 5) encoding function defined as
e(00) = 00000
e(01) = 11011
e(11) = 11100
e(10) = 00101
i) Find minimum distance.
ii) How many errors can e detect?
iii) How many errors can e correct?
Answers :
1. i) Minimum distance of e = 2.
ii) Function can detect 1 or 0 errors.
iii) Function can correct 0 errors.
2. i) Minimum distance of e = 2.
ii) Function can detect 1 or 0 errors.
iii) Function can correct 0 errors.
GROUP CODES:
An (m, n) encoding function e : Bm Bn m n is called a
group code if range of e is subgroup of Bn . i.e. (Ran.(e), ) is a
group. Since Ran.(e) Bn and if (Ran.(e), ) is a group then
Ran.(e) is a subgroup of Bn .
If an encoding function e : Bm Bn m n is a group code,
then the minimum distance of e is the minimum weight of a non zero
codeword.
Example 5 : Show that an (3, 7) encoding function e : B3 B7
defined by
e(000) = 0000000
e(011) = 0111110
e(001) = 0010110
e(101) = 1010011
e(010) = 0101000
e(110) = 1101101
e(100) = 1000101
e(111) = 1111011
is a group code. Hence find minimum distance.

233

Solution :

Let
x 0 = 0000000
x1 = 0010110
x 2 = 0101000
x 3 = 0111110

x 4 = 1000101
x 5 = 1010011
x 6 = 1101101
x 7 = 1111011

Ran.(e) = x 0 ,x1,...,x 7

x 0 x 0 x 0 , x 0 x1 x1 , x 2 x 7 1010011 x 5 like this we can

compute and this we will present in table.

The composition Table is,

x0

x1

x2

x3

x4

x5

x6

x7

x0

x0

x1

x2

x3

x4

x5

x6

x7

x1

x1

x0

x3

x2

x5

x4

x7

x6

x2

x2

x3

x0

x1

x6

x7

x4

x5

x3

x3

x2

x1

x0

x7

x6

x5

x4

x4

x4

x5

x6

x7

x0

x1

x2

x3

x5

x5

x4

x7

x6

x1

x0

x3

x2

x6

x6

x7

x4

x5

x2

x3

x0

x1

x7

x7

x6

x5

x4

x3

x2

x1

x0

Like in Example 4 we can verity that (Ran.(e), ) is group and

Ran.(e) B7 .
Ran.(e) is subgroup of B7 .
e : B3 B7 is a group code.
The minimum distance of a group code is the minimum
weight of non zero code word.
Consider
w x0 0 ,
w x1 w x 4 3 ,
w x2 2 ,
w x5 4 ,

w x3 w x 6 5 ,

Minimum distance = 2.

w x7 6 .

Example 6 : Show that an (2, 5) encoding function e : B 2 B5

defined as
e(00) = 00000
e(10) = 10101
e(01) = 01110
e(11) = 11011
is a group code. Hence find minimum distance and also find
how many errors can e detect?

234

Solution :
x 0 00000 , x1 01110 , x 2 10101 , x 3 11011

Ran.(e) = x 0 ,x1,x 2 ,x 3
The composition Table

x0

x1

x2

x3

x0

x0

x1

x2

x3

x1

x1

x0

x3

x2

x2

x2

x3

x0

x1

x3

x3

x2

x1

x0

Addition is associative

(Ran.(e), ) is associative.

We can see that the first row is same

as heading row.

x 0 is identity element. Also x 0 x 0 x 0 , x 01 x 0 .

x 2 x 2 x 0 . x 2 1 x 2 so on. i.e. inverse of each element exists
which is itself.

Ran.(e), is a group and since Ran.(e) B5 .

Ran.(e) is subgroup of B5 .
e : B 2 B5 is a group code.
Consider,
w x 0 0 , w x1 w x 2 3 , w x 3 4 .

The minimum distance of a group code is the minimum

weight of nonzero code word.

Minimum distance = 3.
Here k + 1 = 3, k = 2.

e can detect 2 or less than 2 errors.

i.e. e can detect 0, 1 or 2

errors.
Check your progress :
1. Show that an (2, 4) encoding function e : B 2 B 4 defined by
e(00) = 0000

e(01) = 0011

e(11) = 1110

e(10) = 1101

is a group code.

235

12.9 LET US SUM UP

In this chapter we have learned that

The product of G1 and G 2 denoted as G1 G 2 .

Homomorphism of a group, its property and types of
homomorphism.

Isomorphism of a group.

Automorphism of a group.

Cyclic group and its generators.

Cosets and normal sub-group and Quotient group.

12.10 UNIT END EXERCISE

Q.1 Define normal subgroup and give one example.
Q.2 If ( C, +) be a group, f:CC define by f(Z) =
for every Z
C, being conjugate of Z, then show that f is Automorphism.
Q.3 Show that A = ({0, 1, 2, 3, 4, 5}, +6) is cyclic.
Q.4 Show that multiplicative group G = {1, -1, i, -i } is cyclic.
Q.5 Let (Z, +) be the group of integers and N = {3n / n Z} then N
is a normal subgroup of Z.
Q.6 If G = {1, -1, i, -i } is a group and G= ({0, 1, 2. 3}, +4) is
another group then show that gG is isomorphism to G.
Q.7 If f : R R defined by f(x) = log(x) for every x R+ then
show that f is isomorphism. Where R+ and R are multiplicative
group.
Q.8 Prove that all finite group of order 2 are isomorphism.
Q.9 Mapping f : G G defined by f(x) = x-1, for all x G on a
group (G, *) is an automorphism if and only if (G, *) is abelian.
Q.10 Show that the group ({0, 1, 2, 3,..n-1}, tn ) is a cyclic
group.
Q.11Show that (Un, . ) is a cyclic group of nth roots of unity under
multiplication.
Q.12 If H is subgroup of G and if x G implies that x2 H, then
prove that H is a normal subgroup of G.

236
0
1
Q.13 Compute
0

1
1
0
0

0
1

0
0

1
1

1
0

1
0
1
0

0
1
.
1

Q.14 Find weights of the given words a) 001110, b) 0000, c)

111, d) 100100110.
Q.15 Find the distance between x and y
i) x = 00111101, y = 00110010
ii) x = 1010001100, y = 0000111100
Answers :
1
1
13
1

0
1
1
0

0
1
0

14 a) 3

b) 0

15. i) 4

ii) 4

c) 3

d) 4

12.11 REFERENCES FOR FURTHER READING

1. University Algebra by N.S. Gopalkrishnan.
2. Contemptary Algebra by Gallian.
3. Discrete mathematics by Kenneth and Rosen.
4. Discrete mathematical structures by Kolman, Busby and Ross.
5. Modern Algebra by Vasishtha.

237

13
RINGS
Unit Structure
13.0 Objectives
13.1 Introduction
13.2 Algebraic structures with Binary Operation
13.3 Rings
13.4 Integral domain
13.5 Fields
13.6 Ring of homomorphism
13.7 Ring of isomorphism
13.8 Let us sum up
13.9 Unit end exercise
13.10 References for further reading

13.0 OBJECTIVES :
After going through this chapter students will be able to:
Algebraic structures with two binary operations.
Definition of ring and its property.
Zero divisor and integral domain.
Fields.
Ring of homomorphism.
Ring of isomorphism.

13.1 INTRODUCTION :
Groups were studied in the previous chapters, and the
definition of group involves a single binary operation with respect to
addition or multiplication. The distributive laws interlink the two
operators addition and multiplication. This leads us to the study of
one such algebraic system equipped with two binary operations
called as rings.

238

Ring is the second algebraic system. The abstract concept of

Rings has its origin from the set of integers. The algebra of rings
follows the pattern already laid out for group. Only difference is
algebraic structures with two binary operations.

13.2 ALGEBRAIC STRUCTURES WITH TWO BINARY

OPERATIONS.
An algebraic structure is a nonempty set together with one or
more binary operation on that set.
Addition and multiplication are both binary operations on the
set R of real numbers is called algebraic structure with two binary
operations. It is denoted by ( R,+, . ).
13.2.1 Rings
Definition: A ring R is a non-empty set with two binary operations
denoted by + and with respect to the following conditions.
R is an abelian group with respect to +,i.e.,
I. a + (b + c) = (a + b) + c a,b,c R
II. there exists 0 R such that a + 0 = a = 0 + a a R
III. For each a R, there exists a R such that a + (-a) = 0 =
(-a) + a
IV. a + b = b + a a, b R
R is a semi group for i.e., a.(b.c) = (a.b). c a, b, c R
Multiplication distributes over assition, i.e.,
I. a.(b + c) = a.b + a.c a,b,c R
II. (b + c).a = b.a + c.a a,b,c R
Note: we write a.b as ab.
Now let us see one example that satisfies the above-described
axioms.
Example 1: Set of even integers is a ring with respect to usual
addition and multiplications of integers.
Solution: Let E be the set of even integers i.e.,
E = {2x : x Z}
Let a, b E where a = 2m and b = 2n
a+b = 2m + 2n = 2(m+n) E ( m+n Z)
E is closed with respect to addition.

239

Let a, b, c E where a = 2m, b = 2n, c = 2p

a+(b+c)=2m+(2n+2p)=2m+2(n+p)=2[(m+n)+p]
=(2m+2n)+2p=(a+b)+c
E is associative with respect to addition.
Since 0 Z, O = 2.0 E
Consider a + O = 2m+2.0 = 2(m+0) = 2m = a
O is the identity element in E.
For m Z there exists m Z and 2.(-m) E.
Let a =2.(-m)
Consider a + (-a) = 2m +2.(-m)
= 2(m+(-m))
= 2.0
= O =(-a) +a
(-a) is the inverse of a.
Inverse exists for each element in E.
For a, b, E then a + b = 2m + 2n = 2(n+m) (sum of integers is
commutative)
= 2n + 2m
=b+a
E is commutative with respect to addition.
(E, +) is an abelian group.
Consider a, b, c Z where a = 2m,b = 2n, c = 2p
a(bc)= 2m(2n.2p) =2m,(4np) = 8 mnp
(ab)c = (2m.2n.)2p =(4mn)2p = 8 mnp
a(bc) = (ab).c a,b,c E
E is associative with respect to multiplication.
Consider a.(b+c) = 2m.(2n 2p)
= 2m. 2n + 2m.2p
= a.b + a.c
Similarly, (b + c).a = b.a + c.a
Distributive laws hold in E.
Hence (E, +, .) is a ring.
Example 2: Show that the set of all rational numbers is a ring with
respect to ordinary addition and multiplication.
Solution: Let Q be the set of all rational numbers.
1) (Q, +) is abelian.

240

Closure: Let a, b Q then a + b Q

because sum of two rational numbers is a rational number.
Associative: Let a, b, c Q then (a + b) + c = a + (b + c)
because associative law for addition holds.
Existence of Identity: 0 Q and 0 + a = a + 0 = a for every a Q
i.e. 0 is additive identity in Q.
Existence of inverse: for every a Q, -a Q and a + (-a) = 0
Hence, additive inverse in Q exists for each element in Q.
Commutative: Let a, b Q then a + b = b + a
because addition is commutative for rational.
2) (Q, .) is a semi group.
Closure : Since the product of two rational numbers is a
rational number.
a, b Q then a .b Q
Associativity: Multiplication in Q is associative.
3) Multiplication is left as well as right distributive over addition in
the set of rational numbers. i.e.
a.(b + c) = a.b + a.c
(b + c).a = b.a + c.a for every a, b,c Q .
Hence, (Q, +, .) is a ring.
13.3.1

Ring with unity

A ring need not have an identity under multiplication, when
a ring other than {0} has an identity under multiplication; we say
that the Ring is with unity.
Definition: R is called a ring with unity element if there exists 1 R
such that a1 = a = 1a for all a 0 R.
Note: A Ring with unity contains at least elements 0 and 1.
Commutative Ring
In a ring, multiplication need not be commutative, when it sis, we
say that the ring is commutative.
Definition: A Ring R is said to be commutative if ab = ba a, b R.
Example 3: Let the addition and multiplication in Q 2 be defined
as x = a + b 2 and y = c + d 2

241

Q ( 2 ) x + y (a + c) + (b + d)
xy = (ac + 2bd) + (ad + bc)
Solution:

2 is a commutative ring with unity.

Q( 2 ) = {a + b 2 / a , b Q}
X = a + b 2 and y = c + d 2 Q( 2 )
x + y = (a + c ) + (b + d)
belong to Q

2 since a + c and b + d

x+yQ 2.
Q 2 is closed with respect to addition.
For x = a + b 2 , y = c + d 2 z = e + f 2 where a,
b, c, d, e, f Q 2
We have, x + (y+z) = a + b 2 + ((c + d 2 ) + (e + f 2 ))
= a + b 2 + (c + e + (d + f) 2 )
= (a + c + e) + (b + d + f) 2
= ((a + c) + e) + ((b + d) + f) 2
= ((a + c) + (b + d) 2 ) + (e + f 2 )
= (x + y) +
Q 2 is Associative with respect to Addition.
Since 0 is a rational number 0 + 0 2 Q 2
Consider (a +b 2 ) + (0 + 0 2 ) = (a + 0) + (b + 0) 2
=a+b 2
Similarly (0 + 0 2 ) + (a + b 2 ) = a + b 2 .
Hence 0 + 0

is the identity existing in Q 2

For a, b Q, -a, -b Q hence (-a) + (-b) 2 ) Q 2

Consider a + b 2 + ((-a) + (-b) 2 ) = (a + (-a) + b + (-b) 2 ) =
0+0 2
Similarly (-a) + (-b) 2 + (a + b 2 ) = 0 + 0 2 .
Hence (-a) + (-b) 2 is the inverse of a + b 2
Inverse exists for each element in Q 2 .
X = a + b 2 and y = c + d 2 Q( 2 ) where a, b, c, d Q
Consider x + y = (a + c) + (b + d) 2
Since addition of rational numbers is commutative.

242

a + c = c + a and b + d = d + b
x + y = (c + a) + (d + b) 2
= (c + d 2 ) + (a + b 2 )
=y+x
Q 2 is commutative with respect to Addition.
(Q 2 , +) is an Abelian group.
Example 4 : Let (G, *) be an arbitrary commutative group and Hom
G be the set of all homomorphisms from (G, *) onto itself. Then
show that (Hom G, +, .) is a ring with unity, where the operation +
defined by
(f + g)(a) = f(a)*g(a), a G, for every f, g Hom G, and
denotes the functional composition.
Solution:
Closure : For every f, g Hom G, and a, b G,
(f, g)(a * b) = f(a * b) * g(a * b)
= (f(a) * f(b)) *(g(a) * g(b))
=(f(a) * g(a) * f(b) * g(b))
= (f + g)(a) * (f + g)(b),
So that the sum f + g Hom G.
Associative : For every f, g, h Hom G, and a G,
We have ((f + g) + h) (a) = (f + g)(a) * h(a)
= ((f(a) * g(a)) * h(a)
= f(a) * ((g(a) * h(a))
= f(a) * (g + h)(a)
= (f + (g+h))(a).
Thus (f + g) + h = f + (g + h).
Existence of identity : For every f Hom G, there exists
constant mapping Z which map all elements of G on e, the identity
of (G, *) such that
(f +Z)(a) = f(a) * Z(a) = f(a) * e = f(a).
Thus f +Z = f Z is an identity in Hom G, that is, the
mapping Z in Hom G is the Zero element.
Existence of inverse: For every f Hom G, -f Hom G
defined by (-f)(a) = f(a)-1 , such that , For every a G,
(f +(-f))(a) = f(a) * f(a)-1 = e = Z(a).

243

Which implies that f +(-f) = Z, therefore inverse also exists.

Commutative property: For every f, g Hom G, a G, we
have
(f + g)(a) = f(a) * g(a) = g(a) * f(a) = (g +f)(a).
Thus (f +g) = f + g,
Hence (Hom G, +) is commutative group.
Similarly we prove that (Hom G, ) is a semi-group with
identity.
Now to prove that (Hom G, +, ) is a ring with unity there
remains to show that is distributive over +.
f (g + h)(a) = f(g + h) a) = f( g(a) * h(a)) = f( g(a)) * f( h(a))
= (fg)(a) * (fh)(a).
Therefore f (g + h) = (fg) + (fh), similarly, we can prove
right distributive law.
Thus (Hom G, +, .) is a ring with unity.

13.4

ZERO DIVISORS

There are some properties, which are not true in a general

ring. We know that product of two integers is zero, if one among
them is zero, but this may no longer be true in any ring R of 2 x 2
0 0
0 10 0 ; even through
matrices we have
0 00 0 0 0

0 0
0 1 are non-zero and their product is zero in R.

0 10 0

Definition: Let R be a ring and a R, b R both are non-zero but

their product
ab = 0 . Then we say that a, b are zero divisors.
13.4.1 Integral Domain
Definition: A commutative Ring, with unity is an integral domain if
it has no zero divisors and it is donated by the symbol.
For example: The Ring of integrals, rational Numbers, and real
numbers and complex numbers is all integral domain.

244

13.5 FIELD
Definition: A commutative Ring R with unit element i 0 in which
every non-zero element has an inverse with respect to multiplication
am called a field.
Example 5: Set of Gaussian integers is an integers domain but not a
field.
Solution: Set of Gaussian integers Z(i) = {a + ib/a, bZ)}
Let x , y Z(i)
Where x = a + ib and y = c + id where a, b, c, d Z
x + y = (a + c) + (b + d) I = ai + ib1 = a + c and b1 = b + d Z.
x . y = (ac bd) + (ad + bc) I
= a2 + ib2 where a2 = ac bd, b2 = ad + bc Z.
+, . are binary operation in Z(i).
Since the element of Z(i) are integers,
We have that
1. Addition and multiplication are commutative in Z(i)
2. Addition and Multiplication are associative in Z(i)
3. Multiplication is distributive over addition in Z(i)
Clearly, zero element 0 = 0 + 0i Z and unit element 1 = 1 + 0i Z
Further, for every x = a + b Z(i), x, y = 0 x = 0 since x , y are
integers.
Z(i) is without zero divisors.
Hence Z(i) is an integral domain
3
4
Let m34iZ i and n i
25
25
So that m.n9 / 2516 / 25i 12 /2512 / 25i
But n Z(i), because 3/25 and 4/25 Z.
So every non-zero element of Z(i) is not invertible
Hence Z(i) is not a field.

13.6 RING HOMOMORPHISM

In groups, one way to discover information about a group is
to examine its interaction with other groups by way of
homomorphism. Now we show that just as a group homomorphism
preserves the group operation, a ring homomorphism preserves the
ring operations.

245

Definition: A ring homomorphism f from a ring R to another ring R1

is a mapping from R to R1 that preserves the two ring operations;
that is, for all a, b in R
f (a + b) = f (a) + f (b)
f (ab) = f (a) f (b).
a b

/ a, b, c Z . Prove or disprove that the

0 c

a b
map : IR Z defined by
a is a ring homomorphism.
0 c

Example: Let R

a b

Solution: : IR Z defined by
a .
0 c
a

Let A 1 1 and B 2 2
0 c1
0 c2
a b a b
( A B) 1 1 2 2
0 c1 0 c2

a a
1 2
0

b1 b2

c1 c2

= a1+ a2
= ( A) ( B)
a b a b
( AB) 1 1 2 2
0 c1 0 c2
a a
1 2
0

a1b1 b1c2

c1c2

= a1 a2
( A) ( B)

a b

a is a ring homomorphism.
0 c

13.7 ISOMORPHISM
Definition: A homomorphism f:R R1 is called an isomorphism if,
f is both one-one and onto mapping.

246

Propertied of homomorphism: Let f : R R1 be a homomorphism

of a ring R into the ring R1 and 0 R, 01 R1 be the zero element of
R and R1 then
f (0) = 01
f (-a) = - (a) a R
f (a-b) = (a) (b) a , b R
Example :- Consider the sings .

/ a, b IR and

b a

S =

a b
is a ring isomorphism.
b a

show that the map (a bi )

Solution :- : C M 2 [ IR ]

a b
(a bi)
.
b a
To show that is homomorphism
A = a+bi
B = c+di

(A+B) = [a+c+bi+di]
bd
ac
=

(b d ) (a c)
= (a b) d (c di)
ad bc
ac bd
( A B)

(ad bc) (ac bd )

a b c d
=

b a d c
= ( A). ( B)
To show that is 1-1
( A) ( B)
(a bi ) (c di )
a b c d
b a d c

a = c, b = d a+bi = c+di
To show that is onto
a b
For any
( M 2 ( IR)
b a
There exists at b, c C such that
a b
(a bi)

b a
a b
Hence, the map (a bi )
is a ring isomorphism.
b a

247

13.8 LET US SUM UP

In this chapter we have learned

Algebraic structures with two binary operation.
Definition of ring.
Commutative ring and ring with zero divisor.
Integral domains and fields.
Ring homomorphism and isomorphism.

13.9 UNIT END EXERCISE

Q.1 Prove that the set I of all integers with ordinary addition and
multiplication as the compositions forms a ring.
Q.2 Show that the set of number given by x +y 5 , where x and y
are integers is a ring with ordinary addition and multiplication as the
two compositions.
Q.3 If E denotes the set of all even integers, then prove that {E, +, .}
ab
is a commutative ring, where a.b =
and + is the usual addition.
2
Q.4 Show that the set of number of the form x +y
rational numbers is a field.

2 , x and y are

Q.5 Show that Z[ 5 ], the set of complex numbers x +y

x, y are integers, ia an integral domain.

5 where

Q.6 Let R is ring with unity e. f : Z R is a mapping defined by

f ( x ) xe x Z . Prove that f is ring of homomorphism.
Q.7 Let f be the function from the integer Z onto the even integers
given by f(x) = 2x for all
x Z. Prove that f is not a homomorphism.

13.10 REFERENCES FOR FURTHER READING

1. University Algebra by N.S. Gopalkrishnan.
2. Contemptary Algebra by Gallian.
3. Discrete mathematics by Kenneth and Rosen.
4. Discrete mathematical structures by Kolman, Busby and Ross.
5. Modern Algebra by Vasishtha.

248

14
RECURRENCE RELATION
Unit Structure
14.0 Objectives
14.1 Introduction
14.2 Series
14.3 Sequences
14.4 Fibonacci
14.5 Generating functions
14.6 Recurrence relations
14.7 Applications of recurrence relations
14.8 Let us sum up
14.9 Unit end exercise
14.10 References for further reading

14.0 OBJECTIVES
After going through this chapter you will be able to:
Series and sequences.
Generating function.
Recurrence relation.
The first order linear homogeneous recurrence relations.
The second order homogeneous linear recurrence relations.
The non-homogeneous relations.
The method of generating functions.
Applications.

14.1

INTRODUCTION:

We all know that the mathematical induction is a proof

technique that verifies a formula or assertion by inductively
checking its validity for increasing values of n. In a similar way, a
recurrence relation is a counting technique that solves an

249

enumeration problem by recursively computing the answer for

successively larger values of n.
The concept of a generating function is one of the most useful
and basic concepts in the theory of combinatorial. The power of the
generating function rests upon its ability nor only to solve the kinds
of problems have we considered so far but also to aid us in new
situations where additional restrictions may be involved.

14.2

SEQUENCES:

A sequence is an ordered list of objects. A sequence is

denoted by a n , where a n represents nth term of the sequence

n . If the list terminates after some steps then we say sequence

is finite otherwise it is called as an infinite sequence.

Example:
(1) 3, 4, 5, 6, 7, 8, 9 is a finite sequence, in this a1 = 3 , a 2 = 4 and so
on.
(2) 1, 4, 9, 16, 25, is an infinite sequence, in this a1 = 1 , a 2 = 4 ,
a 3 = 9, ...

(3) -1, 1, -1, 1, is also an infinite sequence a1 = 1 , a 2 = 1 , a 3 = 1

(4) 1, 3, 7, is an infinite sequence.
In example (1) we can see that a 2 = a1 +1, a 3 = a 2 +1 and so on i.e.
a n+1 = a n +1 , where a1 = 3 and n 7 . Similarly in (4) we have,
a n+1 = 2 a n +1 , where a1 = 1 .
A formula, like above is called as recursive formula, where
next term depends on previous term. A recursive formula must have
a starting value (i.e. a1 ).
But in example (2), we have a1 = 12 , a 2 = 2 2 , a1 = 32 and so on
i.e. a n = n 2 means value of 'a'n+1 does not depend on 'a'n such a
formula is called as Explicit formula. Similarly in (3), a n = 1n
value of a n it is position number.
The set corresponding to a given sequence is the set of all
distinct elements of a given sequence.

250

It can be finite or infinite.

For e.g. (1) for a n = 1n , corresponding set = {-1, 1}.
(2) for a n = n + 1 , corresponding set = {2, 3, 4,}
The difference between set and sequence is, in a set order of
the elements is not important but in a sequence order of the elements
is important.
A set is called countable if it is elements can be arranged in
order first, second, third etc. i.e. it is the set corresponding to some
sequence for example, set of Natural numbers, set of rational etc.
A set which is not countable is called as an uncountable set.
For example, set of Real Numbers.
Check your progress :
1. Write a formula for nth term and identify it is recursive or explicit.
1 1 1
2 3 4

(a) 1, , , , ...

(b) 1, 0, 1, 0, 1, 0, ...

(c) 3, 6, 9,
(e) 5, 25, 125,

(d) 2, 5, 10, 17, 26

14.3 SERIES
An expression of the form a1 + a2 + a3 + + an +
. which is the sum of the elements of the sequence {an} is
called a series. If the series contains a finite number of elements, it is
called finite series, otherwise called an infinite series.
If Sn = a1 + a2 + a3 + + an , then Sn is called the sum of
n terms of the series and is denoted by the Greek letter sigma .
n

Thus S n an .
i 1

Example 1: Find the sum of first 20 natural numbers.

Solution: To find sum of first 20natural numbers.
i.e Sn = 1 + 2 + 3 +4 +..+20.
Here first term = a = 1 and the common difference = d =1
Bye arithmetic progression,
Sn =
=
= 10[ 2 + 19]= 210.

251

Check your progress:

1. Find the sum of the series 2+4+6+8+10++38.
2. Find the sum of the series 3+5+7+9+11++53.
3. Find the sum of the series 3+7+11+15+.+73.

14.4 FIBONACCI SEQUENCE:

The Fibonacci sequence is a set of numbers that starts with a
one and a zero, followed by a one , and proceed based on the rule
that each number is equal to the sum of the preceding two numbers
is called Fibonacci number and the sequence obtained is called
Fibonacci sequence.
If Fibonacci sequence is denoted by F(n), where n is the first
term in the sequence, the following equation obtains for n=0,where
first two term are defined as 0 and 1 by convention.
F(0) = 0, 1, 1, 2, 3, 5, 8, 13, .

14.5 GENERATING FUNCTIONS

Now we see some important polynomial expansions, which
are often used in this chapter.
Polynomial Identities
1 x n 1
1 x x 2 x3 ........ x n
1 x
1
2.
1 x x 2 x 3 .........
1 x
n n
n
n
3. (1 x) n 1 x x 2 x3 ........ x n
1 2
3
n

1.

Definition: Let a0, a1, a2, be a sequence of real numbers. The

function

f x a 0 a1x a 2 x 2 ai r i
i 0

is called the ordinary generating function or generating function

for the given sequence.
Example 2: Find the generating function for the binomial theorem .
Solution: For any n Z+
n n n
n
(1 x) n x x 2 ................ x n .
0 1 2
n
n
So (1 x) is the generating function for the sequence

252

n n n
n
, , ,,
0 1 2
n
f ( x) (1 x) n is the generating function for a r C (n, r ) , the number

of ways to select an r-subset of n-set.

Example 3: Find the coefficient of x5 of generating function
(1 2 x)7 .
Solution : Using the generating function for Maclaurin series.
7
r
(2 x)
r 0 r

We write (1 2 x)7 (1 y ) 7

where y = -2x.

Consequently, the coefficient of x5 is

7
5
5 7 5 1
(2 x) (1)
(32)
5
5
11
(32) 14, 784.
5

Check your progress:

1. Find the generating function for Maclaurin series. [ hint
f ( x) (1 x) n ]
(1 x n1 )
2. Find the generating function for
.
(1 x)
3. Determine the coefficient of x8 of generating function
1
.
( x 3)( x 2)5

14.6

RECURRENCE RELATION

A recurrence relation is a recursive formula that counts the

number of ways to do a procedure involving n objects in terms of the
number of ways to do it with fewer objects. That is if a, is the
number of ways to do the procedure with r objects, for r=0, 1,
2,, then a recurrence relation is an equation that expresses an as
some function of preceding ak s, k < n. A formal definition is given
after the following example.
A boy has a staircase of n stairs to climb. Each step it takes
can cover either one stair or two stairs. Find the number of different
ways for the boy to climb staircase.

253

We solve this using recurrence relation.

n

(n-1)

an1
(n-2)

an2

3
n=3

2
n=2

a1
Fig 14.1
Let an be the number of different ways the boy can climb a staircase
with n stairs.
Thus
a1=1,
[one step can be climbed in only one way]
a2=2,
[1+1 that is in two steps or both stairs in single step]
a3=3,
[1+1+1, 1+2 and 2+1]
etc.
to compute an, the boy can reach n th stair with or without using n-1
th stair using n-1 th stairs he has an-2 ways without using n-1 th stair
(as the has to use n-2 th stair) he has an-2 ways.
Thus
an = an-1 + an-2.
These numbers are well known Fibonacci numbers. Where n th term
is some of previous two terms such dependency gives us recurrence
relation which can be formally defined as:

254

A recurrence relation is a relation among the sequence (an), n=0, 1,

2,.., of the form f(n, a1, a2,.,an ) = 0 for all n k.
Three simple example could be
an an-1 an-2 = 0, n 2,
(1)
an - an-1 = n, n 1,
(2)
an a1an-1 a2an-2 -..- an-1 a1 = 0, n 3.
(3)
is a linear homogeneous recurrence relation with constant
coefficients,
is a linear non-homogeneous recurrence relation with constant
coefficients,
is a non-linear recurrence relation, and a formal definition is below.
Let k Z+ and Cn ( 0), Cn-1, Cn-2,, Cn-k ( 0) be real number.
If an, for n 0, is a discrete function , then
Cnan + Cn-1an-1 + Cn-2an-2 +.+ Cn-kan-k = f(n), n k ,
Is a linear recurrence relation with constant coefficient of
order k. when f(n) = 0 for all n 0, the relation is called
homogeneous; it is non homogenous.
14.6.1 THE FIRST ORDERED LINEAR RECURRENCE
RELATION
A linear homogeneous recurrence relation of the form,
an = an-1
is called the first order homogeneous linear recurrence relation,
where is a constant.
Since an depends only on its immediate predecessor, the relation is
said to be first-order.
Now the above relation may be defined as the first order linear
homogenous recurrence relation with constant coefficients.
General solution
The general solution of the recurrence relation
an = an-1

(4)

where is a constant, n 1, and a0 = A , is unique and is given by

an = A n
(5)
The expression a0 = A is called the initial condition .
Thus the solution an = A n , n 1, defines a discrete function whose
domain is the set N of all natural numbers.
Example 4: Solve the recurrence relation an+1 = 3an , n 0, a0 = 5.

255

Solution:
The first four terms of this sequence are
a0 = 5,
a1 = 3a0 = 3(5) = 15,
a2 = 3a1 = 3(3a0) = 32 (5) , and
a3 = 3a2 = 3(32(5)) = 33(5).
By using induction, we can easily verify that, the general solution is
an = 5(3n) for each n 1.
In the general solution, the value of an is a function of n and
there is no longer any dependence on prior terms of the sequence,
once the value of a0 is known.
Example 5: Solve the recurrence relation an = 7aa-1 , where n 1 and
a2 = 98.
Solution:
This is an alternative from of the recurrence relation aa+1 = 7
an for n 0 and a2 = 98.
From the above formula, we have the general solution an = aa (7) n
Since a2 = 98 = a0 (72) . This gives a0 = 2.
Hence the general solution an = 2(7) n , n 0 , is unique.
Example 6: A bank pays 6% annual interest on savings,
compounding the interest monthly. If a person deposits 1000/- , how
much will this deposit be worth a year later?
Solution:
We solve the above problem using the recurrence relation
concept as follows.
Let Pn be the deposit at the end of n months.
Given
P0 = 1000,
The rate of interest per annum is 6%, so the monthly rate =
6/12% = 0.005.
Now the recurrence relation for the deposit is,
Pn+1 = P0 + 0.005Pn = PN(1.005)
Where 0.005Pn is the interest earned on Pn at the end of n the month.
The general solution for the recurrence relation is
Pn = P0 (1.005)n

256

Consequently the deposit at the end of 12 months is

P12

= P0(1.005)12
= 1000(1.005)12
= 1061.68

14.6.2 Special Cases

Conversion of Non Liner to Linear: In a linear relation there are
no products such as anan-1, (see equation (1) in the definition of
recurrence relation at the beginning of this chapter) which appears in
the non-linear relation, an+1 3anan-1 = 0. However , there are times
when a nonlinear recurrence relation can be transformed into a linear
one by a suitable algebraic substitution.
Example 7: Find a8 if a2n+1 = 6a2n , where an > 0 for n 0 , and
a0 = 2.
Solution:
Given the recurrence relation
a2n+1 = 6a2n, where an > 0 for n 0 , and a0 = 2 (6)
is not a linear one in an
But we transform (6) into linear with the substitution.
bn = a2n ,
Then the new relation
bn+1 = 6bn for n 0, and b0 = 4,
(7)
is a linear relation
Now the general solution for the new recurrence
relation (7) can be computed using the general solution
method discussed in this section 3.1.
Comparing (6) with equation (4), we observe that
= 6 and the general solution is
`bn = ()n b0 = (6)n 4.
Therefore
an = 2( ) n for n 1.
Consequently
a8 = 2( ) 8 = 2592.
Example 8: Find a12, if an+12 = 5an2 , where an > 0 for n 0 and
a0 = 2.
Solution:
This is a non-linear recurrence relation. This can be converted into
linear by letting bn = an2. This gives the new relation bn+1 = 5bn for
n 0, and b0 = 4 which is linear.

257

The general solution is bn = 4(5) n.

Therefore, an = 2( ) n for n 0.
Hence a12 = 2( ) 12 = 31,250.6
14.6.3 Non Homogeneous Relations: The general first order linear
recurrence relation.
With constant coefficients has the form an+1 + can = (n), n 0
where c is a constant and (n) is a function on the set N of Non
negative integers.
Where (n) = 0 for all n N, the relation is called homogeneous;
otherwise it is called non homogeneous. So far, we have dealt only
with homogeneous relations. Now we shall solve a nonhomogeneous relation.
Example 9: Find a recurrence relation and solve it for the following
sequence.
Solution:
0, 2, 6, 12, 20, 30, 42
Hear a0 = 0, a1 = 2, a2 = 6, a3 = 12, a4 = 20, a5 = 30, a6
= 42, and we observe that
a1 a0 = 2,
a2 a1 = 4,
a3 a2 = 6,
a4 a3 = 8,
a5 a4 = 10,
a6 a5 = 12,
These calculations suggest the recurrence relation.
an an-1 = 2n, n 1, a0 = 0
To solve this relation, we proceed in a slightly different manner from
the method we used in the previous examples. Consider the
following n equation:
a1 a0 = 2,
a2 a1 = 4,
.
.
.
an an-1 = 2n,
By adding the above n equations we get,
a1 a0 = 0 + 2 + 6 + 8 + + 2n
= 2(1 + 2 + 3 + + n)
= 2[n(n+1)/2] = n2 + n

258

Therefore the general solution is

an = n2 + n.
With variable coefficients
We have seen and discussed various types of recurrence
relations, which are with constant coefficients. Now we see an
example of the type with variable coefficients in this section First
order recurrence relations.
Example 10:Solve the relation an = n.an-1 , when n 1 and a0 = 1
Solution: We solve this problem by proceeding as above but differ
in a small manner.
Given the recurrence relation.
an = n.an-1 , when n 1 and a0 = 1
The first terms of the relation are
a0 = 1,
a1 = 1,a0 = 1,
a2 = 2,a1 = 2.1,
a3 = 3,a2 = 3.2.1,
a4 = 4,a3 = 4.3.2.1,
a5 = 5,a4 = 5.4.3.2.1,
Therefore, an = n! And the solution is the discrete
function an, which counts the number or permutations
of n objects, n 0.
We have seen the same counting principal in chapter 1
of this book in the permutations section.
14.6.5 THE SECOND ORDER LINEAR HOMOGENEOUS
RECURRENCE
RELATION
WITH
CONSTANT
COEFFICIENTS
Hear we talk about the relations of particular case in which k = 2,
discussed in section 3.1, equation (3).
The relation of the from
Cnan + Cn-1an-1 + Cn-2an-2 = 0 , n 2
(8)
Is called the linear recurrence homogeneous relation with constant
coefficients of order 2.
General Solution
As per our discussion 3.1, we found the general solution of
the form
an = crn
(9)
Where c 0, r 0.
Substituting (9) in (8) we get
Cncrn + cn-1crn-1 + cn-2crn-2 = 0
(10)

259

Which becomes
Cnr2 + cn-1r + Cn-2 = 0.
(11)
a quadratic equation of degree two called the characteristic
equation.
Let the roots of the characteristic equation be r1, r2. Then the
following cases arise: roots may be
a. Distinct i.e. r1 r2,
b. Complex Conjugates or
c. Equal.
Note that to determine the unique solution of a second order linear
homogeneous recurrence relation with constant coefficients we need
to have two initial conditions, that is the value of an for at least two
values of n.
Case (I) Roots are Distinct
Let the distinct roots be r1, r2 and r1 r2. Since the
general solution of (8) is
an = crn , where r is the root of equation (10), we have
an r and an = r are both solutions to (8), and
linearly independent.
Therefore the general solution in this case is
an = c1 r + c2 r
Where c1, c2 are arbitrary constants.
Note that the solutions an = r and an = r linearly independent when
the following condition is satisfied.
For c1, c2 R, if c1r + c2r = 0 for all n N, then c1 = c2 = 0.
We may find the unique solution by using the initial conditions and
eliminating the arbitrary constants.
Example 11: Solve the recurrence relation an + an-1 6an-2 = 0,
where n 2 and a0 = -1
and a1 = 8.
Solution:
Given the recurrence relation
an + an-1 6an-2 = 0, where n 2
(12)
n
On substituting the general solution an = cr , c,r 0
and n 2 in (12) we obtain
crn + crn-1 - 6crn-2 = 0
and the corresponding characteristic equation is
r2 + r 6 = 0.

260

Solving the characteristic equation.

r2 + r 6
=0
(r + 3)(r 2) = 0
r = 2, -3.
Since we have two distinct real roots, the general solution is
an = c1(2)n + c2(-3)n
Where c1 c2 are arbitrary constants.
The arbitrary constants are determined using the initial conditions as
follows:
-1
= a0
= c1(2)0 + c2(-3)0
= c1 + c2
(13)
8
= a1
= c1(2)1 + c2(-3)1
= 2c1 - 3c2
(14)
Solving (13) and (14) we find the values of the arbitrary constants as
c1 = 1, c2 = -2.
Case (II) Complex Roots
Let the complex roots be r1,r2. Since the general solution of (8) is
an = crn, where r is the root of equation (10),
an = r and an = r are both solutions to (8), and
linearly independent as in the case(I). Therefore the
general solution in this case is
an = c1r + c2r
Where c1, c2 are arbitrary constants.
Since r1, r2 are complex roots (complex conjugates), let
these roots be
r1= a + ib and r2 = a ib. now re writing the general
solution we get
an = c1(a + ib) + c2(a ib)
(15)
We may find the unique solution by using the initial
conditions and eliminating the arbitrary constants. And the linearly
independence as is case (I)
Note that the solution contains no complex numbers.
This is demonstrated in the following example.
Example 12: Solve the recurrence relation an = 2(an-1 an-2), where
n 2 and a0 = 1, a2 = 2.

261

Solution:
Let the general solution be
an = crn, for c, r 0.
Substituting the solution in the given recurrence
relation we get the characteristic equation.
r2 -2r + 2 = 0.
Solving the above equation we obtain the roots are
r = 1 i.
Now we have two independent solutions (1 + i)n and (1 i)n.
Therefore the general solution is
an = c1(1 + i)n + c2(1 i)n.
Where c1, c2 are arbitrary complex constants.
Using preliminary concept of complex analysis we rewrite the
general solution without complex numbers and eliminate the
arbitrary constants using the initial conditions as follows.
1 + i = (cos( / 4) + i sin( / 4))
1+i=

and
(cos( / 4) + i sin( / 4)) =

(cos( / 4) - i sin( / 4))

These yields
an = c1(1 + i)n + c2(1 i)n
= c1( (cos( / 4) + i sin( / 4)))n + c2( (cos( / 4)
- i sin( / 4)))n
`=
( )n(c1(n(cos(n / 4) + i sin(n / 4))) + c2((cos(n
/ 4) - i sin(n / 4))))
=
( )n(k1 cos(n / 4) + k2 sin(n / 4)).
Where k1 = c1 + c2 and k2 = (c1 c2)i.
1 = a0 = k1 cos0 + k2 sin0 = k1.
2 = a2 = (cos( / 4) + k2 sin( / 4)) = 1 + k2.
Therefore
k1 = 1, k2 = 1.
The solution for the given initial conditions is then given by
a2 = ( ) n (cos(n / 4) + sin(n / 4)), n 0.
Case (III) Repeated Roots
Let the r1, r2 and r1 = r2, in this case we say that the root of
multiplication 2. Since the general solution of (8) is an = crn, where r
is the root if equation (10), we have an = are both solution to (8).
But these two solutions are not linearly independent and we say

262

these are dependent solution since one solution is obtained from the
other by just multiplying with 1, i.e. multiples of each other.
To find the general solution in this case is we need one more
independent solution.
Let the independent solution be (n) rn where (n) is not a
constant. Substituting this into the given relation yields a general
solution of the from
an = c1rn + c2nrn,
And we see this in the example below.
Example 13: Solve the recurrence relation an+2 = 4an+1 4an, where
n 0 and a0 = 1, a1 = 3.
Solution:
First we find the characteristic equation of the given relation
by letting the general solution be an = crn, c, r 0. This yields the
characteristic equation
r2 4r + 4 = 0.
Solving the above equation we find roots, r = 2,2 the repeated
roots. Now the two dependent solutions are 2n and 2n.
Let the independent solution be
(n) 2n
Where (n) is not a constant.
(n + 2)2n+2 = 4 (n + 1)2n+1 - 4 (n)2n
or
(n + 2) = 2 (n + 1) - (n).
Assuming that the general solution is (n) = an + b, for
arbitrary constants a, b, with a 0. Hear we choose a = 1, b = 0.
Therefore we find that (n) = n satisfies the above equation.
So n2 is a second independent solution.
Now the general solution is of the form
an = c12n + c2n2n
Using the initial condition, a0 = 1, a1 = 3 we find that
an = 2n + (1/2)n2n , n 0.
n

14.6.6
THE
RELATION

NON-HOMOGENEOUS

RECURRENCE

A general form of the non homogeneous first and second

order relations are

263

an + Cn-1an-1 = (n),
n 1, Cn-1 0.
(16)
an + Cn-1an-1 + Cn-2an-2 = (n),
n 2, Cn-2 0.
(17)5
Where Cn-1, Cn-2 are constants and (n) is not identically zero.
General solution
There is no general method for solving non-homogeneous
relation, for certain functions (n) we shall find a successful
technique.
A Special Case
When Cn-1 = -1 in equation (16) we have
a1 = a0 + (1)
a2 = a1 + (2) = a0 + (1) + (2)
a3 = a2 + (3) = a0 + (1) + (2) + (3)
. . .
an = a0 + (1) + (2) + (3)+ + (n) = a0 +

i1 f i .
i n

We can solve this type relation in terms of n, if we can find s

suitable summation formula for i1 f i .
i n

Example 14:Solve the recurrence relation an an-1 = 3n2, where n

1 and a0 = 7.
Hear (n) = 3n2, so the general solution is

Solution:
an

= a0 + i1 f i .
i n

= 7 + 3 i1 i 2 .
i n

n n 12n 1
2
Method of undetermined coefficients
This relies on associated homogeneous relation obtained by
replacing (n) with zero.
= 7

Let a n ( h ) denote the general solution for the associated homogenous,

relation and a n ( p ) be a solution of the given non-homogeneous
relation, the term a n ( p ) is called the particular solution. Then
an = a n ( h ) + a n ( p ) is the general solution of the given relation. We use
the form (n) to suggest a form for a n ( p )
Here is the procedure to find general solution in which (n) = krn, k
is a constant. Consider the non-homogeneous first order relation.
an + Cn-1an-1 = krn , n 1, Cn-1 0.

264

If rn is not a solution of the associated homogeneous relation.

an + Cn-1an-1 = 0, then
a n ( p ) = Arn,
Where A is a constant. When rn is a solution of the associated
homogenous relation.
a n ( p ) = Bnrn,
For B a constant.
Now consider the case of the non-homogeneous second order
relation
an + Cn-1an-1 + Cn-2an-2 = krn,
n 2, Cn-2 0
Here we find that
( p)
a. a n = Arn, A is a constant, if rn is not a solution of the
associated homogenous relation;
b. a n ( p ) = Bnrn, B is a constant, if a n ( h ) = c1rn + c2rn, where r1
r ; and
c. a n ( p ) = Cnrn, C is a constant, when a n ( h ) = (c1 + c2n)rn.
Example 15: Solve the recurrence relation an 3an-1 = 5(7n), where
n 1 and a0 = 2.
Solution:
Given the non homogenous relation
an 3an-1 = 5(7n),
(18)
and
(n) = 5(7)n .
The corresponding homogenous relation is
an 3an-1 = 0.
(19)
Solving (19), the general solution for the homogenous
relation is
a n ( h ) = c(3n).

Let the particular solution of (18) be

a n ( p ) = c(3n).
A is a constant and
(n) = krn = 5(7)n
Now we have
a n ( h ) = A (7n)

265

Placing the particular solution in (18), we find that

A(7n)- 3A(7n-1) = 5(7n), n 1

7A 3A = 5(7)

And

A = 35/4,

a n ( p ) = (35 / 4)7n

= (5 / 4)n+1 , n 0.
The general solution is an = c(3n) + (5 + 4)7n+1
Using the initial condition a0=2 now, we find the value of the
arbitrary constant c,
2= a0
= c + (5 / 4) (7)

c
= - 27 / 4.
Finally the unique general solution is
an =(5/4)7n+1 (1/4)(3n+3), n 0.
Example 16:
Solve the recurrence relation an 3an-1 = 5(3n), where
n 1 and a0 = 2.
Solution: Given the non homogenous relation
an 3an-1 = 5(3n),
(20)
And
(n) = 5(3)n
(21)
The corresponding homogenous relation is
an 3an-1 = 0.
Solving (21), the general solution for the homogenous
relation is
a n ( h ) = c(3n)
Here a n ( h ) and (n) are not linearly independent. As a
result a n ( p ) of the form Bn(3n). Substituting the
particular solution in the given relation we get
Bn(3n) = 3B(n-1) (3n-1) = 5(3n)
Or
Bn B(n-1) = 5.
Therefore B = 5.
Hence
an = a n ( h ) + a n ( p ) = (c + 5n)3n , n 0.
With a0 = 2, the general solution is
an = (2 + 5n)3n

266

14.7 THE METHOD OF GENERATING FUNCTION

The title of this topic itself indicates that the assistance of
Generating Function in solving recurrence relation is involved.
Now we demonstrate the procedure to solve a given recurrence
relation with the help of generating function in the following
example in a systematic procedure.
Example 17:Solve the recurrence relation
an+2 5an+1 + 6an = 2, n 0.
a0 = 3,a1 = 7
Step 1
Multiply the given relation by x n2 , because n 2 is largest
subscript in the relation. This gives us
Step 2
Sum all the equations represented by the result in step (1) and we get

n0 an2 x n2 5 n0 an1xn2 6 n0 an xn2 2 n0 xn2

Step 3
In order to have each of the subscripts on a match the corresponding
exponent on x, we rewrite the equation in step (2) as

n0 an2 x n2 5 x n0 an1x n1 6 x2 n0 an x n 2 x 2 n0 x n

Step 4

Let f x n0 an x n be the generating function for the solution.

The equation in step (3) now takes the form

Or

2x2
2
f
x

a
x

5
x

f
x

6
x

f
x

0 1
0
1 x

f x3 7 x5 x f x3 6 x 2 f x

2x2
1 x

Step 5

Solving for f x we have

2x2
1 5 x6 x f x 38 x
1 x
3 11x 10 x 2

1 x
2

267

Form which it follows that

3 11x 10 x 2
f x
1 5 x 6 x 2 1 x

3 5 x 1 2 x
1 3x 1 2 x 1 x

3 5x
1 3x1 x
Partial fraction decomposition gives us
2
1
f x

1 2 x 1 x

2 3 x x
n0

n0

Consequently, an = 2(3n) + 1, n 0,
Check your progress:
1. The number of bacteria in a culture is 1000, and this number
increases 250% every two hours. Use a recurrence relation to
determine the number of bacteria present after one day.
2. If a person invests Rs.100 at 6% interest compounded quarterly,
how many months, must he wait for his money to double? (He
cannot withdraw his money before the quarter is up.)
3. A person invested the stock profits he received 15 years ago in an
account that paid 8% interest compounded quarterly. If his
account now had Rs.7218.27 in it, what was his initial
investment?
4. Using generating functions solve the recurrence relation
an 3an-1 = n, n 0, a0 = 1.
5. Solve the recurrence relation
an+2 = an+1 + an, n 0, a0 = 0, a1 = 1
6. Solve the recurrence relation
an+2 4an + 3an = -200, n 0, a0 = 3000 , a1 = 3300.

268

14.8

LET SUM UP:

In this chapter, we have learnt the following.

Sequences and series.
Generating function and Example base on Binomial and
Maclaurin series.
A recurrence relation is a relation among the sequence (an), n =
0, 1, 2, , of the form (n, a1, a2, ,an) = 0 for all n k.
A linear homogenous recurrence relation of the form,
an = an-1
is called the first order homogenous linear recurrence relation,
where is a constant.
The general solution of the recurrence relation an = an-1 where
is a constant, n 1 , and a0 = A, is unique and is given by an =
An .
A general form of the non homogenous first and second order
relation are
an + Cn-1an-1 =
, n 1, Cn-1 0
an + Cn-1an-1 + Cn-2an-2 =
,
n 2,
Cn-2 0,
Where Cn-1Cn-2 is constants and f n is not identically zero.
Solving various recurrence relations and the method of
generating function.

14.9 UNIT END EXERCISE:

Determine the coefficient x7 of generating function (1 3x) 9 .
Solve ar ar 1 ar 2 30 given a0 20, a1 5.
Solve ar 2ar 1 ar 2 0 .
Solve ar 2ar 1 ar 2 3 with a0 2, a1 5.
Solve the recurrence relation an+an-1-6an-2=0, n2, and a0=-1 and
a1=8.
6. Solve the recurrence relation an+2=4an+1-4an for n2, a0=1, a1=3.
7. Solve the recurrence relation an-3an-1=5(7n) for n1 and a0=2.
8. Solve the recurrence relation an-3an-1=5(3n) for n1 and a0=2.
9. Solve the relation an+2-5an+1+6an=2, n0, a0=3, a1=7.
10. Determine the coefficient x6 of generating function (1 5 x)8 .
1.
2.
3.
4.
5.

269

14.10 REFERENCES:
1. Discrete structures by Liu.
2. Discrete mathematics its Application, Keneth H. Rosen TMG.
3. Discrete structures by B. Kolman HC Busby, S Ross PHI Pvt.
Ltd.
5. Discrete mathematics, schaums outlines series, seymour Lip
Schutz, Marc Lipson, TMG.