1/11/2023

Graph Theory
Lecture 3

Dr. Ilyas Fakhir

 Representation of Graphs inside a Machine

1/11/2023
• In order to classify the various computer models of graphs, the
three following principles can be distinguished:
 The natural way to represent a non-simple graph is the
Adjacency matrix. This is a square matrix of order n, M = (mij),
where mij is the number of edges having xi & xj endvertices in
G. This implementation of graph takes memory space of order
n2, where n is the number of vertices. So any algorithm over
graphs modeled after adjacency matrix has O(n2) time
complexity.

mij =  1 if edge exist between vertex xi and vertex x j

0 otherwise.
Representation of Graphs inside a Machine

1/11/2023
(Contd…)
 A second principle for implementing a graph is to give for each
vertex its “neighborhood”. The most classic way to implement
this principle is showing the neighbors as adjacency list. List
of neighbors N(x) for each vertex x. The memory space need for
this type of implementation is of order n + m, n vertices and m
edges. Such model is coherent with linear processing.
 If in an algorithm the entry to the graph is done by the edges
and not directly by the vertices, as in the above models, a third
implementation principle consists of giving a list of edges with,
for each of them, its endvertices. Such a list may be linked,
which makes it possible to modify the graph being processed.
The memory space necessary for this type of model is minimal
since it is of the order of the number of edges m of the graph
(the isolated vertices do not appear in this implementation,
since they are not endvertices of an edge).
Representation of Graphs inside a Machine

1/11/2023
(Contd…)
 Weighted Graph

1/11/2023
• Weighted Graphs are often considered, that is graphs with
values, integer or real, positive or not, associated with the edges.
• Formally, we have a graph G = (X, Y ) with a mapping v : E → R.
• When a weighted graph is a simple graph, which is often the
case, its computer model is generally a matrix, such as the
adjacency matrix, but with entries being the values of the edges
under consideration.
• The matrix M = (v(xixj)), where 1 ≤ i, j ≤ n, with mapping v
extended by stating: v(xixj) = ∞ when i ≠ j and xixj ∉ E, v(xixj) = 0
when i = j. This matrix is symmetric.
Incidence Matrix

1/11/2023
• If the edge set of undirected graph G is E(G) = {al, a2, ... , ae},
• then the incidence matrix NG of G is the v x e matrix with entries
nij, such that

nij =  1 if vertex xi is incident with edge a j

0 otherwise.

nij = 2 if vertex xi has self-loop

• The adjacency and incidence matrices depend on the orderings
chosen for V(G) and E(G); they are not unique, but vary only by
row and/or column permutation.
• Column represents edges and Rows represents vertices
 1/11/2023
Isomorphism in Graph
• The simple graphs G1 = (V1, E1) and G2 = (V2, E2) are
isomorphic if there is a bijection (a one-to-one and onto
function) f from V1 to V2 with the property that a and b are
adjacent in G1 if and only if f(a) and f(b) are adjacent in G2, for
all a and b in V1.
{a,b}∈E1⇔{f(a),f(b)}∈E2.
• Such a function f is called an isomorphism.
• In other words, G1 and G2 are isomorphic if their vertices can
be ordered in such a way that the adjacency matrices MG1 and
MG2 are identical.

7
 1/11/2023
Isomorphism (Contd…)

• In other words, two graphs G1 and G2 are said to be

isomorphic (to each other) if there is a one-to-one
correspondence between their vertices and between their
edges such that the incidence relationship (that is the
adjacency of vertices) is preserved.
• Such graphs will have the same structures differing only in
the way their vertices and edges are labeled or only in the
way they are represented geometrically. For many purposes,
we regard them as essentially the same. When G1 and G2 are
isomorphic we write G1 ≡ G2 .

8
 A B P Q

1/11/2023
D C S R
• Observe that both graphs have four vertices and four edges.
• The one-to-one correspondence between the vertices are:
A↔P; B↔Q; D↔S and C↔R.
• The one-to-one correspondence between the edges are:
(A,B) ↔ (P,R); (A,D)↔(P,S); (B,C)↔(Q,R); (C,D)↔(S,Q).
• These represent one-to-one correspondence between the edges of
the two graphs which preserve the incidence relationship, i.e.,
the adjacent vertices and adjacent edges in the first graph
correspond to adjacent vertices and edges in the second graph.
Thus they are isomorphic. 9
 1/11/2023
Graph Isomorphism-Example

• EG: Prove that

is isomorphic to

• First label the vertices:

2 2
1 3 1 3
5 4
5 4
10
 1/11/2023
Graph Isomorphism-Example
 1/11/2023
Example
Show that the multiplicative group G consisting of three cube
roots of unity 1,ω,ω2 is isomorphic to the group G′ of residue
classes under addition of residue classes (mod3).
1/11/2023
 1/11/2023
Subgraphs

• LetG and H are two graphs. They are said to be sub

graphs if the following conditions hold
 All the vertices and edges of H are in G
 Each edge of H has the same end vertices in G as in H.
• Inother words, if H is a graph with vertex set V(H) and
the edge set E(H) and G is graph with vertex set V(G)
and the edge set E(G) then H is a subgraph of G
whenever V(H) is subset of V(G) and E(H) is subset of
E(G).

14
 1/11/2023
Proper subgraph

•A subgraph H is said to be proper subgraph of G if G

has at least one vertex which is not in H.

A A
B B

C C

D D

E
15
 1/11/2023
• The following results are immediate consequences of the
definition of Subgraph
 Every graph is a subgraph itself
 Every simple graph of n vertices is a subgraph of the
complete graph Kn
 If K is subgraph of H and H is subgraph of G then K is
subgraph of G
 A single vertex in a graph G is a subgraph of G
 A single edge in a graph G together with its end vertices
is a subgraph of G

16
Edge-disjoint and Vertex-disjoint

1/11/2023
subgraphs
• Let G be graph. Let K and H are two graphs of G. Then
 K and H are said to be edge-disjoint if they do not have any
common edge
 K and H are said to be vertex-disjoint if they do not have any
common edge and any common vertex.
• Notice that the edge-disjoint subgraphs may have common
vertices. Subgraphs that have no vertices in common cannot
possibly have edge in common, i.e., two vertex-disjoint
subgraphs must be edge-disjoint but the converse is not
necessarily true.

17
 1/11/2023
• Example of edge-disjoint but not vertex-disjoint subgraphs

A
C A
B

B D

E` E
E
Subgraph K Subgraph H
D
Graph G

18
 1/11/2023
Operations on graphs

• Union of two graphs

 Let G1(V1, E1) and G2(V2, E2) be two graphs. Then the
graph whose vertex set is V1V2 and the edge set
E1 E2 is called the union of G1 and G2 and it is
denoted by G1  G2
G1 G2 = (V1 V2, E1 E2)

19
 1/11/2023
• Intersection of two graphs
 Let G1(V1,E1) and G2(V2,E2) be two graphs. Then the
graph whose vertex set is V1∩V2 and the edge set
E1∩E2 is called the intersection of G1 and G2 denoted
by G1∩G2 is
G1∩G2 = (V1∩V2, E1∩E2), Provided V1∩V2 ≠ϕ
• Ring sum
 Consider a graph whose vertex set is V1UV2 and the
edge set E1 E2, where  is the symmetric difference
of E1 and E2 then G1 G2=(V1UV2, E1 E2) is called
the ring sum of G1 and G2 .
 Note: Symmetric difference means
E1E2 = (E1 – E2)U(E2 – E1)=(E1UE2) –(E1∩E2)

20
• The following consequences of the above definition of union,

1/11/2023
intersection and ring sum
 G1U G2 = G2U G1
 G1 G2 = G2 G1
 If G1 and G2 have no edge in common (edge-disjoint), then G1∩
G2 is a null graph and G1 G2 = G1U G2
 If G1 and G2 have no vertex in common (vertex-disjoint), then
G1∩ G2 does not exist
 Sometimes, we also write G1 + G2 for G1U G2 .
 Note that G1,G2 ⊆ G1U G2 and G1∩ G2 ⊆ G1,G2.
 G – U := G[V \U, E], G – F := (V, E \F ), G + F := (V, E U F ).
If U = {u} or F = {e} then we simply write G – u, G – e and G + e
for G – U, G – F and G + F , respectively.
 The complement of G, denoted by G or GC, is defined as the
graph (V; V1V2\ E). In particular, G + G is a complete graph,
and G = (G + G) – E. 21
 1/11/2023
V2 V2
V1 V1 V6

V3

V3
V5 V4
V4 G2
G1
V1 V2 V6
V2
V1 V6 V1 V2

V5 V4 V3
V5
V4 V3 V4 V3 G1G2
G1UG2 G1∩G2
22
Complement of Graph

1/11/2023