UNIT-6:Graph Theory

 Basic concepts
 Representation of Graphs
 Isomorphism and Sub graphs
 Spanning trees
 Planar graphs
 Multi graphs and Euler Circuits
 Hamiltonian graphs
 Chromatic number
Basic concepts and notations.
 A Graph G is a pair of sets (V, E)
 where V = A set of vertices (nodes) and
 E = A set of edges (lines)
 V(G) = Set of vertices in G.
 E(G) = Set of edges in G.
 V(G)  = Number of vertices in graph G = Order of G.
 E(G)  = Number of edges in graph G = Size of G .
Types of graphs
 Non Directed Graph (Undirected graph): The elements of E
are unordered pairs (sets) of vertices. In this case an edge {u, v} is
said to join u and v or to be between u and v.
 Directed Graph: In a digraph the elements of E are ordered pairs
of vertices. In this case an edge (u, v) is said to be from u to v.

 Loop: An edge drawn from a vertex to itself.

 Multi Graph: If one allows more than one edge to join a pair of
vertices, the result is then called a multi graph.

Simple Graph: A graph with no loops and no parallel

edges.
Definitions
 Degree: Degree of a vertex in an undirected graph is the number
of edges incident with it, except that a loop at a vertex contributes
twice to the degree of that vertex. The degree of the vertex ‘v’ is
denoted by deg(v).
 In-degree and Out-degree: In a digraph, the number of edges
incident to a vertex is called the in-degree of the vertex and the
number of vertices incident from a vertex is called its out-degree.
 The in-degree of a vertex ‘v’ in a graph G is denoted by deg+(v) .
 The out-degree of a vertex v is denoted by deg -(v).
 A loop at a vertex in a digraph is counted as one edge for both
in-degree and out-degree of that vertex.
 Neighbors: If there is an edge incident from u to v, or incident on
u and v, then u and v are said to be adjacent ( neighbors).
 (G) = minimum of all the degrees of vertices in a graph G.
 (G) = Maximum of all the degrees of vertices in a graph G.
Regular Graph: In a graph G, if (G) = (G) = k i.e., if each
vertex of G has degree k, then G is said to be a regular graph of
degree k (k-regular).
 Ex: Polygon is a 2-regular graph .
 Ex: A 3-regular graph is a cubic graph.

Complete Graph: A simple non directed graph with ‘n’

mutually adjacent vertices is called a complete graph on ‘n’ vertices
and may be represented by Kn.
Note: A complete graph on ‘n’ vertices has [{n(n – 1)}/ 2]edges, and
each of its vertices has degree ‘n-1’.
Every complete graph is a regular graph.
The converse of the above statement need not be true.
 Cycle Graph: A cycle graph of order ‘n’ is a connected graph
whose edges form a cycle of length n.
 Note: A cycle graph ‘Cn’ of order n has n vertices and n edges.
 Null Graph: A null graph of order n is a graph with n vertices and
no edges.
 Wheel Graph: A wheel graph of order ‘n’ is obtained by adding a
single new vertex (the hub) to each vertex of a cycle graph of
order n-1.
 Note: A wheel graph Wn has ‘n’ vertices and 2(n-1) edges.
 Bipartite Graph: A Bipartite graph is a non directed graph
whose set of vertices can be partitioned in to two sets M and N in
such a way that each edge joins a vertex in M to a vertex in N.
 Complete Bipartite Graph: A complete Bipartite graph is a
Bipartite graph in which every vertex of M is adjacent to every
vertex of N.
 If M = m and N  = n then the complete Bipartite graph is
denoted by Km , n. It has ‘m n’ edges.
 The number of edges in a bipartite graph is  (n2/4).

Degree Sequence: If v1, v2, ……, vn are the vertices of a

graph G, then the sequence{d1, d2,….., dn} where di =
degree of vi is called the degree sequence of G.
Usually we order the degree sequence so that the degree
sequence is monotonically decreasing.
First theorem on graph theory

 Sum of Degrees Theorem: If V = {v1, v2, ….., vn} is the vertex set
of a non directed graph G then

  deg (vi) = 2.  E
n
 iProof:
=1 When the degrees are summed, each edge contributes a
count of one to the degree of each of the two vertices on which
the edge is incident.
 Hence, the theorem follows.
Corrollories

 Cor. 1 If G is a digraph, then

  deg+(vi) =  deg –(vi) =  E

n n
 Cor.2. An undirected
i=1 graph hasi=1
an even number of vertices of odd
degree. (See Text book for proof)
 Cor.3 If G is a k– regular graph , then k. V = 2.  E .

 Cor.4 In a non directed graph G, If k = (G) and m = (G) then

 k. V  2. E  m .V

Examples

 Ex. A non directed graph contains 16 edges and all vertices are of
degree 2. Find the number of vertices in G ?
 Solution: By Sum of degrees theorem, if degree of each vertex is k,
then
 k. V = 2. E
  2.  V = 2.(16)
   V = 16
Example

 Ex. A simple non directed graph G contains 21 edges, 3 vertices of

degree 4 and the other vertices are of degree 2. Find the number of
vertices in the graph G ?
 Solution: Let  V = n . By Sum of degrees theorem,

  deg (vi) = 2.  E
n

 i=1 3.(4) + (n – 3 ). 2 = 2.(21)
  n = 18.
Ex.What is the number of vertices in an undirected connected graph
with 27 edges, 6 vertices of degree 2, 3 vertices of degree 4 and
remaining vertices of degree 3?
 Solution: Let  V = n . By Sum of degrees theorem,

  deg (vi) = 2.  E

  6.(2) + 3.(4) + (n – 9 ). 3 = 2.(27)

  n = 19.
Ex. If a simple non directed graph G contains 24 edges and all vertices
are of same degree then find the number of vertices in G?

 Solution: By Sum of degrees theorem, if degree of each vertex is k,

then
 k. V = 2. E
  k.  V = 2.(24)
   V = 48/k ( k = 1,2,3,4,6,8,…)
  V(G) = 48, 24, 16, …
Ex. What is the largest possible number of vertices in a graph G, with
35 edges and all vertices are of degree at least 3 ?.

 Solution: By Sum of degrees theorem, if degree of each vertex is 

k, then
 k. V  2. E
  3. V  2.(35)
   V  (70/3 = 23.33..)
   V  23
  The largest possible number of vertices = 23
Example

 Ex. Let G be a simple graph with n vertices. Then show that the
number of edges in G is less than or equal to {n.(n – 1)} / 2.

 Solution: In a simple graph, each edge correspond to a distinct pair of

vertices.
 The number of ways to choose a pair of vertices, out of n vertices =
 C(n , 2) = n. (n – 1 )
 2
 The maximum number of edges possible in G = {n.(n – 1)} / 2.
  The number of edges in G is less than or equal to {n.(n – 1)} / 2.
Ex. Which of the following degree sequences represent a simple non directed
graph
a) {2, 3, 3, 4, 4, 5} b) {2, 3, 4, 4, 5}
c) {1, 3, 3, 4, 5, 6, 6} d) {1, 3, 3, 3}
 Solution: a) Consider the degree sequence {2, 3, 3, 4, 4, 5}
 Here, we have 3 vertices with odd degree.
 But, By sum of degrees theorem, An undirected graph should
contain an even number of vertices of odd degree.
  The degree sequence cannot represent a simple non directed
graph.

 b) Consider the degree sequence {2, 3, 4, 4, 5}

 Here, we have a vertex with degree 5.
 But, A simple non directed graph of order 5 cannot have a vertex
with degree 5.
  The degree sequence cannot represent a simple non directed
graph.
Contd.,

 c) Consider the degree sequence {1, 3, 3, 4, 5, 6, 6}

Here, the order of the graph is 7 and we have 2 vertices with degree 6.
 Since these two vertices are adjacent to all other vertices of the graph, a
vertex with degree 1 does not exist
  The degree sequence cannot represent a simple non directed graph.

 d) Consider the degree sequence {1, 3, 3, 3 }

Here, the order of the graph is 4 and we have 3 vertices with degree 3.
 Since these 3 vertices are adjacent to all other vertices of the graph, a
vertex with degree 1 does not exist.
  The degree sequence cannot represent a simple non directed graph.
Ex. Show that a degree sequence with all distinct elements cannot
represent a simple non directed graph.

 Solution: Let G = {v1,v2,v3,…,vn}

 The possible degree sequences are {0,1,2,….,n –1} and {1,2,3,…,n}
 In a simple graph of order n, if there is a vertex with degree n – 1
then a vertex with degree 0 does not exist.
 A simple non directed graph of order ‘n’ cannot have a vertex with
degree n.
  The degree sequence with all distinct elements cannot represent
a simple non directed graph.
Representation of Graphs

 Adjacency list: One way to represent a graph with no multiple edges is

to use adjacency lists, which specify the vertices that are adjacent to
each vertex of the graph.
 1) Adjacency matrix: The adjacency matrix of a graph is a
 matrix with rows and columns labeled by the vertices and such that
 its entry in row i, column j is 1 otherwise it is 0
 For instance the following is the adjacency matrix of the graph of
figure
2) Incidence matrix: The incidence matrix of a graph G is
a matrix with rows labeled by vertices and columns
labeled by edges, so that entry for row v column e is 1 if e
is incident on v, and 0 otherwise. As an example, the
following is the incidence matrix of graph of figure
Path Matrix
Isomorphism
 Isomorphic Graphs: Two graphs G and G1 are isomorphic if there
is a function f : V(G)  V(G1) such that
 (i) f is a bijection and
 (ii) for each pair of vertices u and v of G,
 {u, v}  E(G)  {f(u), f(v)}  E(G1)
 i.e.. the function preserves adjacency.
 Note: If G is isomorphic to G1 then
 a)  V(G) =  V(G1)
 b)  E(G) =  E(G1)
 c) The degree sequences of G and G1 are same.
 d) If {v, v} is a cycle in G, then {f(v), f(v)} is a loop in G1, and more
generally, if v0 – v1 – v2 – …. – vk – v0 is a cycle of length k in G, then
f(v0)– f(v1)– f(v2) – … – f(vk) – f(v0) is a cycle of length k in G1.
Isomorphism
 Suppose G and G1 are two graphs and that f : V(G)  V(G1) is a
bijection.
 Let A be the adjacency matrix for the vertex ordering v1, v2, ……,vn of
the vertices of G.
 Let A1 be the adjacency matrix for the vertex ordering
 f(v1), f(v2), ….. , f(vn) of the vertices of G1.
 Then f is an isomorphism from V(G) to V(G1) iff the adjacency
matrices A and A1 are equal.
 Note: If A  A1, then it may still be the case that graphs G and G1 are
isomorphic under some other function.
Isomorphism

 Complement of a graph: The complement of a graph G is the

graph G with the same vertices as G. An edge exists in G iff it does
not exist in G.
 Theorem: Two simple graphs are isomorphic iff their complements are
isomorphic.
 If two graphs are isomorphic, then their corresponding sub graphs are
isomorphic.
 Induced Subgraph: If W is a subset of V(G), then the sub graph
induced by W is the sub graph H of G obtained by taking V(H) = W
and E(H) to be those edges of G that join pairs of vertices in W.
 If G is isomorphic to G1 then G is said to be self
complementary.
Isomorphism
 Ex. If G is self complementary graph with n vertices, then show that
G has {n(n – 1)}/4 edges.
 Ex. If G is self complementary then prove that G has 4k or 4k + 1
vertices where k is some positive integer.
 Ex. How many non isomorphic graphs are there of order 4 and size 2?
 Ex. How many non isomorphic graphs are there of order 8, size 8 and
degree sequence {2,2,2,2,2,2,2,2}.
 Ex. How many non isomorphic graphs are there of order 6, size 6 and
degree sequence {2, 2, 2, 2, 2, 2}.
 Ex. Let Cn be a cycle graph on n vertices. if Cn is isomorphic to Cn
then show that n = 5.
Spanning Trees
 Tree: A connected graph with no cycles is called a tree.
 A tree with ‘n’ vertices has (n – 1) edges.
 A tree with n vertices (n>1) has at least two vertices of degree 1.
 A sub graph H of a graph G is called a spanning tree of G if
i) H is a tree and
ii) H contains all vertices of G
 Note: In general, if G is a connected graph with n vertices and m
edges, a spanning tree of G must have (n – 1) edges. Therefore, the
number of edges that must be removed before a spanning tee is
obtained must be m – (n – 1). This number is called circuit rank of G.
 A non directed graph G is connected iff G contains a spanning tree.
 The complete graph Kn has nn – 2 different spanning trees.
(Caley’s formula)
Minimal Spanning Tree: Let G be a connected graph where each
edge of G is labeled with a non negative cost. A spanning tree T where
the total cost C(T) is minimum is called a minimal spanning tree.
 Kruskal’s Algorithm: (For finding minimal spanning tree of a
connected weighted graph)
 Input: A connected graph G with non negative values assigned to
each edge.
 Output: A minimal spanning tree for G.
 Method: 1) Select any edge of minimal value that is not a loop.
This is the first edge of T(if there is more than one edge of
minimal value, arbitrary choose one of these edges)
 2) Select any remaining edge of G of having minimal value that does
not form a circuit with the edges already included in T.
 3) Continue step 2 until T contain (n – 1) edges when n = |V(G)|
Prim’s algorithm.
 Prim’s Algorithm: (For finding a minimal spanning tree)
 1) Let G be a connected graph with non negative values assigned to
each edge. First let T be the tree consisting of any vertex V1 of G.
 2) Among all the edges not in T, that are incident on a vertex in T and
do not form a circuit when added to T, Select one of minimal cost and
add it to T.
 3) The process terminates after we have added (n – 1) edges
where n = |V(G)|.
 Construction of Spanning Tree
 DFS(Depth First Search)
It uses Stack.
The principle of the algorithm is quite simple: to go
forward (in depth) while there is such possibility, otherwise
to backtrack.
 BFS(Breadth First Search)
It uses Queue.
Breadth-first search starts with given node
Then visits nodes adjacent in some specified order
(e.g.,alphabetical)
Planar Graphs
 A graph or a multi graph that can be drawn in a plane or on a sphere
so that its edges do not cross is called a planar graph.
 Ex : A complete graph on 4 vertices K4 is a planar graph.
 Ex : Tree is a planar graph.
 Map, Connected map : A particular planar representation of a finite
planer multi graph is called a map. We say that the map is connected if
the under lying multi graph is connected.
 Region : A given map (planar graph) divide the plane into connected
areas called regions
 Degree of a region : The boundary of each region of a map consists
of a sequence of edges forming a closed path. The degree of region ‘r’
denoted by deg (r) is the length of the closed path bordering r .
Sum of degrees of regions theorem
 If G is a planar graph with k regions, then the sum of the degrees of
the regions of G is equal to twice the number of edges in G.
i.e.,  deg (ri) = 2. E .
k
 Cor.1 In a planar
i=1
graph G, if the degree of each region is k then
k.R  = 2 .E
 Cor.2 In a planar graph G, if the degree of each region is  k , then
k.R   2 .E
 In particular, If G is a simple connected planar graph ( A planar graph
with no loops and no parallel edges, and degree of each region is  3),
then
3. R   2 . E 
Euler’s formula
 Question: State and prove Euler’s formula for planar graphs.
 Statement: If G is a connected planar graph, then
 V –  E + R  = 2.
Proof: …..by Mathematical Induction
 Theorem: If G is a simple connected planar graph with  E > 1 then,
(a)  E  {3. V – 6}.
(b) There exists at least one vertex v of G such that deg(v)  5
 Theorem: If G is a simple connected planar graph with  v > 3 then,
 R  {2. V – 4}.
Polyhedral graph
 Polyhedral Graph: A connected planar graph is said to be
polyhedral if degree of each region is  3 and deg(v)  3
for all v  G.
i.e., 3. R   2 . E  and 3. V   2 . E .
 For any polyhedral graph
 a) V   (2 + R  )
 2
 b) R   (2 + V  )
 2
 c) (3. R - 6)  E 
Euler Path and Circuit
 Euler Path: An Euler path in a multigraph is a path that includes
each edge of the multi graph exactly once and intersects each vertex
of the multi graph at least once.
 An Euler circuit is an Euler path whose end points are identical.
 A multigraph is traversable if it has Euler path.
 A non directed multi graph has an Euler path iff it is connected and
has zero or exactly two vertices of odd degree.
 Any finite connected graph with exactly two odd vertices is
traversable(Euler path exists). A traversable trail may begin at either
odd vertex and will end at other odd vertex.
 A connected multi graph has an Euler circuit if and only if all of its
vertices are of even degree.
Hamiltonian Graph
 Hamiltonian Graph: A Hamiltonian Graph is a graph with a closed
path that includes every vertex exactly once. Such a path is a cycle
and is called a Hamiltonian cycle.
 An Eulerian circuit uses every edge exactly once but may repeat
vertices , while a Hamiltonian cycle uses each vertex exactly once
(except for the first and last) but may skip edges.
Kurtowski Theorem
 A graph G is not planar iff G contains a sub graph homomorphic to
K3, 3 or K5 .
 Ex. A planar graph contains 25 vertices and 60 edges then find the
number of regions in the graph .
 Ex. What is the maximum number of edges possible in a planar graph
with eight vertices ?.
 Ex. What is the minimum number of vertices necessary for a graph
with 11 edges to be a simple planar graph?.
 Ex. Suppose that a connected planar graph has 20 vertices, each of
degree 3. In to how many regions does a representation of this planar
graph split the plane ?
Examples
 Ex. Let G is a connected planar graph with 35 regions and degree of
each region is 6. Find the number of vertices in G ?.
 Ex. Suppose G is a polyhedral graph with 12 vertices and 30 edges
prove that degree of each region is 3.
 Ex. Show that there does not exist a polyhedral graph with exactly
seven edges.
 Ex. Show that there does not exist a polyhedral graph with exactly 30
edges and 11 regions.
 Theorem: Prove that a complete graph Kn is planar iff n  4.
 Theorem: Prove that a complete Bipartite graph Km, n is planar
iff m  2 or n  2.
Chromatic number
 Vertex coloring: A coloring of a simple graph is the assignment of
color to each vertex of the graph so that no two adjacent vertices are
assigned the same color.
 Chromatic Number: The minimum number of colors needed to
paint a graph G is called the chromatic number of G , denoted by
(G)
 Adjacent Regions: In a planar graph two regions are adjacent if they
share a common vertex.
 Map coloring: An assignment of colors to the regions of a map such
that adjacent regions have different colors.
 A map ‘M’ is n – colorable if there exists a coloring of M which uses n
colors.
 A planar graph is 5 – colorable
Four color Theorem
 Four color Theorem: If the regions of a planar graph are colored so
that adjacent regions have different colors, then no more than 4
colors are required.
i.e., (G)  4.
 Ex. Prove that the chromatic number of a complete graph Kn is n.
 Ex. Prove that the chromatic number of a complete Bipartite graph
Km, n is 2.
 Ex. Prove that the chromatic number of cyclic graph Cn is 2 if n is
even and 3 if n is odd.
 Ex. If every cycle of G has even length then show that its chromatic
number is 2.
 Ex. Prove that the chromatic number of a tree on n vertices is 2.