Graph Isomorphism

† Goal: is it possible to draw 2 graphs

in the same way?
† In chemistry, different compounds
can have the same molecular formula
but can differ in structure
† The graphs of these compounds may
not be drawn in the same way
Graph Isomorphism
† Two identical graphs with geometrical
differences is called isomorphic graphs
† The simple graphs G1=(V1,E1) and
G2=(V2,E2) are isomorphic if there is a one to
one correspondence, that u and v are adjacent
in G1 if and only if u’ and v’ are adjacent in G2
† When two simple graph isomorphic, there is
one to one correspondence between vertices of
two graphs that observes the adjacency
relationship
Graph Isomorphism
† Two graphs are isomorphic iff they
are identical except for their vertex
names.
Graph Isomorphism
3 d c v w

1 2 a b x y
(a) G1 (b) G2 (c) G3

G1 isomorphic with G2, but G1 is not isomorphic with G3

 Graph Isomorphism
a b c d e
a ⎡0 1 1 1 0⎤
z ⎢1 0 1 0 0⎥
b ⎢ ⎥
a v w c ⎢1 1 0 1 0⎥
⎢ ⎥
e d ⎢1 0 1 0 1⎥
e ⎢⎣0 0 0 1 0⎥⎦
c
b d
x y w v z
x y
x ⎡0 1 1 1 0⎤
y ⎢1 0 1 0 0⎥
(a) G1 (b) G2
⎢ ⎥
w ⎢1 1 0 1 0⎥
v ⎢1 0 1 0 1⎥
⎥
⎢
Graph (a) and graph (b) isomorphic z ⎢⎣0 0 0 1 0⎥⎦
Graph Isomorphism

Two graphs are isomorphic

Graph Isomorphism

Tree graphs are isomorphic

Connectivity
Introduction
† Many problems can be modeled with
paths formed by traveling along the
edges of graphs. For instance :
„ determining whether a message can be
sent between two computers using
intermediate link
„ diagnostics in computer networks
„ and so on
Path
† A path is a sequence of edges that begins
at a vertex of a graph and travels along
edges of the graph, always connecting pairs
of adjacent vertices
† Let n be nonnegative integer and G an
undirected graph. A path of length n from u
to v in G is sequence of n edges (x0,x1),
(x1,x2),...,(xn-1,xn), where x0=u and xn=v
† When the graph is simple, we denote this
path by its vertex sequence x0,x1,...,xn
† The path is a circuit if it begins and ends at
the same vertex, that is u = v
Path
Observe graph below:
† path 1, 2, 4, 3 or (1,2), (2,4), (4,3), has length = 3
† path 1, 2, 3, 1 is a circuit and has length = 3

3 2

4
Connectedness in
Undirected Graphs
† Two vertices v1 and v2 are called
connected if there is a path between
them
† An undirected graph is called
connected if there is a path between
every pair of distinct vertices of the
graph
Connectedness in
Undirected Graphs
† Graph below is not connected. Why?

2
5

1 4
6

3 8 7
Subgraph and
Complement of Subgraf
† Suppose G = (V, E) is a graph.
G1 = (V1, E1) is subgraph of G if V1 ⊆ V and
E1 ⊆ E.
† Complement of subgraf G1 to graph G is
graph G2 = (V2, E2) such that E2 = E - E1 and
V2 is set of vertices which E2 members are
incident to it
Subgraph and
Complement of Subgraf

2 2

1 1 1
3 3
3

6 6

4 5 4 5 5

(a) Graph G1 (b) a subgraph (c) complement

of subgraf
 Spanning Subgraph
† subgraph G1 = (V1, E1) of G = (V, E) is called
spanning subgraph if V1 =V (G1 contains all
vertices of G).

1 1 1

2 3 2 3 2 3

4 5 4 5

(a) graph G, (b) spanning subgraf (c) not spanning subgraph

of G of G,
Cut-Set
Cut-set of connected graph G is set of
edges, that if we remove these edges
from G, G be not connected
Thus, cut-set produces 2 connected
components
Cut-Set
† {(1,5), (1,4), (2,4), (2,3)} is cut-set. There are
many cut sets on a connected components
† Set {(1,5), (4,5)} is also cut-set, {(1,2), (1,4),
(1,5)} is cut-set, {(5,6)} also cut-set,
† But {(1,5), (4,5), (3,4)} is not cut-set since its
subset, {(1,5), (4,5)} is cut-set.

1 5 1 5

4 6 4 6

2 3 2 3
Euler Path and Circuit
Introduction
Konigsberg bridge problem
Introduction
† Problem : Traveling Graph was modeled by
across every bridge Euler :
without crossing C
any bridge more
than once. Is there
a simple circuit in A D
this multigraph
that contains every
edge?
B
Euler Path and Circuit
† An Euler path in G is a simple path
containing every edge of G
† An Euler circuit in a graph G is a simple
circuit containing every edge of G
† A graph containing Euler circuit is called
Eulerian Graph. A graph containing Euler
path is called Semi-Eulerian Graph
Euler Path and Circuit
† Euler path on graph (a) : 3, 1, 2, 3, 4, 1
† Euler path on graph (b) : 1, 2, 4, 6, 2, 3, 6, 5, 1, 3
† Euler circuit on graph (c) : 1, 2, 3, 4, 7, 3, 5, 7, 6, 5, 2,
6,1

2 2 2 3
1 1

(a) (b) (c) 5

3

4 1 4

3 4 5 6 6 7
Euler Path and Circuit
† Euler Circuit on graph (d) : a, c, f, e, c, b, d, e, a,
d, f, b, a
† Graph (e) and (f) do not have Euler path nor Euler
Circuit

a
d b 1 2 a b
(d) (e) (f)
3

e c 4 5 c d e

f
THEOREM
† A graph is called Semi-Eulerian Graph
iff the graph contains two vertices
with odd degree
† A graph is called Eulerian Graph iff
each vertex has even degree
Examples
† Mohammed’s scimitars
Hamilton Path and Circuit
Introduction
† The term Hamiltonian circuit comes from a
game, called Icosian puzzle, invented by
Irish mathematician Sir William Rowan
Hamilton. It consisted of a wooden
dodecahedron. The 20 vertices of the
dedocahedron were labeled with different
cities in the world.
† The object of the puzzle was to start at city
and travel along the edges of the
dodecahedron, visiting each of the other 19
cities exactly once, and back at the first city
Introduction
Hamilton’s Dedocahedron
Hamilton Path and
Circuit
† Hamilton circuit is a circuit passing by
each vertex just once. A graph containing
Hamilton circuit is called Hamiltonian
graph.
† Hamilton path is a path passing by each
vertex just once. A graph containing
Hamilton path is called Semi-Hamiltonian
graph
Hamilton Path and
Circuit
(a) Semi-Hamiltonian graph (path : 3, 2, 1, 4)
(b) Hamiltonian graph (circuit : 1, 2, 3, 4, 1)
(c) Non-Hamiltonian graph, non-Semi-Hamiltonian graph

1 2 1 2 1 2

4 3 4 3 4 3
(a) (b) (c)
Theorem
† Necessary (but not sufficient) conditions for existence
of Hamilton circuits were found by Gabriel A. Dirac
(1952) and Oystein Ore (1960)
† Dirac’s Theorem
If G is simple graph with n vertices with n ≥ 3 such
that the degree of every vertex in G at least n/2, then
G has a Hamiltonian circuit
† Orec’s Theorem
If G is a simple graph with n vertices with n≥3 such
that d(u) + d(v) ≥ n for every pair of non adjacent
vertices u and v in G, then G has a Hamilton circuit
Theorem
† Several things about Hamiltonian
graph
† Every complete graph is Hamiltonian graph
† On a complete graph G with n vertices (n ≥
3), there are (n - 1)!/2 Hamilton circuit
† On a complete graph G with n vertices (n ≥
3 and n is odd), there are (n - 1)/2
mutually exclusive circuits.
† On a complete graph G with n vertices (n ≥
4 and n is even), there are (n - 2)/2
mutually exclusive circuits
Graph Applications
†Shortest Path
†Traveling Salesman
Problem
†Chinese Postman Problem