Graphs

[Module – 5]

WCE - SY (CSE) – 1CI 202: Data Structures

Course Teacher: Dr. Smriti H. Bhandari
 Graphs
 Basic Terminology
 Representation of Graphs
 Adjacency Matrix
 Adjacency List
 Adjacency Multilist
 Traversals
 Depth first
 Breadth first
 Minimum Spanning Tree
2
 Basic Concepts
 Non-linear data structure

 Definition
 Graph is
 a collection of nodes, called vertices and
 A collection of segments or lines called edges/arcs
connecting pairs of vertices

3
 Basic Concepts (2)
 Formal definition
 Graph G is a pair (V,E) where
 V is a set of nodes called vertices
 E is a set of pairs of vertices each representing an
edge

4
 Basic Concepts (3)
 Directed Edge
 Ordered pair of vertices (u,v)
 First vertex u is the origin
 Second vertex v is the destination

 Undirected Edge
 Unordered pair of vertices (u,v)

5
 Basic Concepts (4)
 Directed graph or digraph
 All the edges are directed

 Undirected graph
 All the edges are undirected

6
 Applications
 Electronic circuits
 Printed circuit boards
 Integrated circuits
 Transportation and communication networks
 Highway network
 Flight network
 Computer Networks
 Local area network
 Internet
 Databases
 Entity-relationship graph
 Modeling any sort of relationships
 Between components, processes, people, concepts etc.

7
 Example – Flight route network
 Edge represents flight route between two
airports

8
 Terminology
 End vertices or end points of an edge
 U and V are end points of edge a

 Edges incident on the vertex

 a, d and b are incident on V

 Adjacent vertices
 Vertex v is adjacent to vertex u iff (u,v)  e
 U and W are adjacent

 Degree of a vertex
 Number of edges incident to it or number of
adjacent vertices
 X has degree 4

 Loops
 Edge having identical end points
 i forms loop at Z

 Isolated Node/Vertex
 A node which is not adjacent to any other node
 Degree of isolated node is zero
 S is an isolated node 9
 Terminology – (2)
 Path
 A sequence of vertices v1 ,v2 ,. . .vk such that vi+1 is adjacent to vi
for i = 1 .. k – 1
 Simple path
 A path with no repeated vertices (i.e. all vertices are distinct)
 Cycle
 A path except that the last vertex is same as the first vertex
 Simple Cycle
 A simple path except that the last vertex is same as the first vertex
 Cyclic graph
 Graph containing a cycle
 Acyclic graph
 Graph which does not contain a cycle
10
 Terminology – (3)
 Connected graph
 A graph G is connected if and only if there is a simple
path between any two nodes in G

 Complete graph
 A graph G is said to be complete if every node u in G
is adjacent to every other node v in G.
 Always connected
 A complete graph with n nodes will have n(n-1)/2
edges.

11
 Terminology – (4)
 Strongly Connected Graph
 A directed graph G is said to be strongly
connected, if for each pair u,v of nodes in G
there is a path from u to v and there is also a
path from v to u

 Unilaterally Connected Graph

 Graph G is said to be unilaterally connected if
for any pair u, v of nodes in G there is a path
from u to v or a path from v to u
12
 Terminology – (5)
 Densely Connected Digraphs
 The ratio of number of edges to number of
nodes is large

 Sparsely Connected Digraph

 The ratio of number of edges to number of
nodes is small

13
Terminology – (6)
 n 
  di 
e   i 1 
2
where
e  number of edges
n  number of vertices
d i  degree of vertex i

14
 Terminology – (7)
 Tree graph
 A connected graph T without any cycles is called a tree
graph or tree
 There is a unique simple path P between any two
nodes u and v in T
 Null graph
 Graph containing only isolated nodes
 Graph containing an empty set of edges
 Order of Graph
 The number of vertices

15
 Terminology – (8)
 Labeled Graph
 A graph G is said to be labeled if its edges are
assigned data

 Weighted Graph
 A graph G is said to be weighted if each edge
e in G is assigned a nonnegative numerical
value w(e) called the weight or length of e.

16
 Terminology – (9)
 A graph G is called a simple
graph if it satisfies the following
conditions:

 There are no loops in the graph G.

This means, that there does not
exist an edge in E of form (Vi, Vi)
for any Vi in V.

 There is not more than one edge

joining any pair of vertices Vi and,
Vj in V. This means, that there
does not exist more than one edge
in E of the form (Vi, Vj).
17
 Representation of Graph
 Matrix Representation

 List Representation

 Multilist Representation

18
 Adjacency Matrix Representation
 Consider a graph G with set of vertices V and
set of edges E. If there are N nodes in V, for
N > = 1, the graph G may be represented by
an adjacency matrix which is a table with N
rows and N columns where

A(i, j )   1
0
if and only if there is an edge (Vi , Vj )
otherwise

19
 Adjacency Matrix
 If G is an undirected graph, then
adjacency matrix A of G will be symmetric
matrix
 i.e. for every i and j, A(i,j) = A(j,i)

20
Adjacency Matrix – Example 1
A B C D

0 1 1 1 A

1 
0 0 0 B
A
1 0 0 1
 
1 0 1 0 C
21
Adjacency Matrix – Example 2
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

22
 Adjacency Matrix – Observations
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

i th row in the adjacency matrix is determined by

the edges which start from the vertex Vi.
23
 Adjacency Matrix – Observations
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

The number of elements in the i th row whose

value is 1 is equal to the out degree of the node
Vi. 24
 Adjacency Matrix – Observations
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

The number of elements whose value is 1 in j th

column is equal to the in degree of vertex Vj.

25
 Adjacency Matrix – Observations
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

If there are loops at each node but no other

edge, then the adjacency matrix is the identity
or the unit matrix. 26
 Adjacency Matrix – Observations
X Y Z W

Y X
0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 
0 0 1 0 W

For a null graph which has no edges, the

adjacency matrix has all its elements as zero.
Therefore, its adjacency matrix is a null matrix. 27
Adjacency Matrix – Example 3

A B C D E F G

A 0 5 0 0 0 0 0

B 0 0 6 5.5 0 0 0

C 0 0 0 0 0 10 0

D 0 0 0 0 7 0 0

E 0 0 0 0 0 7.5 0

F 0 0 0 0 0 0 4

G 0 0 0 0 0 0 0

28
Adjacency Matrix – Example 4

A B C D

A 0 1 0 1

B 1 0 1 1

C 0 1 0 1

D 1 1 1 0

29
 An entry of 1 in the i th row and j th column
of A in the adjacency matrix of a graph
shows that there exists an edge (Vi ,Vj)
that means, there exists a path of length 1
from Vi and Vj

30
 Proposition
 Let A be the adjacency matrix of a graph
G. Then Ak(i,j) i.e the ij entry in the matrix
Ak, gives the number of paths of length K
from vi to vj.

31
 Example
X Y Z W

Y X 0 0 0 1X
1 0 1 
1Y
A
Z W
1 0 0 1 Z
 W
0 0 1 0
Compute A2, A3, A4
32
 X Y Z W
Y X 0 0 1 0 1 0 0 1
1 0 1 
2 3 1 
0 2 2
A2   A 
0 0 1 1 1 0 1 1
Z W    
1 0 0 1 0 0 1 1

•There is a path of length 2 from W to X 0 0 1 1

2 0 2 3
•There are two paths of length 3 from Y
to Z and from Y to W A4  
1 0 1 2
•There are three paths of length 4 from  
Y to W
1 0 1 1
33
 Proposition
 Let us define matrix Bm as below:

B m = A + A2 + A3 + … + Am

 The entry ij of matrix Bm gives the number

of paths of length m or less from node vi to
vj

34
 Path Matrix / Reachability Matrix
 Let G be the simple directed graph with m nodes
v1,v2, … vm. The path matrix or reachability
matrix of G is the m-square matrix P defined as
below:

P(i, j )   1
0
if and only if there is a path from Vi to Vj
otherwise

35
 Proposition
 Let A be the adjacency matrix and let P be
the path matrix of a digraph G.
Then P(i,j) = 1 if and only if there is a
nonzero number in the ij entry of the
matrix
B m = A + A2 + A3 + … + Am

36
 Example
 Using earlier example compute B4 and P

Y X
1 0 2 3 1 0 1 1
5 
0 6 8 1 0 1 
1
B4   P 
3 0 3 5 1 0 1 1
Z W
   
2 0 3 3 1 0 1 1

 Node Y is not reachable from any of the other

nodes
37
 Observation
 Graph G is strongly connected if and only if
the path matrix P of G has no zero entries

38
 Drawbacks of Adjacency matrix
representation
 Static implementation
 A priori knowledge regarding number of
nodes in the graph is needed
 Dynamic changes in the graph are difficult
to model
 Wastage of space if matrix is sparse
 Not possible to store additional information
about the graph
39
 Linked Representation
 A linked structure that can make allocation
and deallocations of nodes dynamically

 Also known as adjacency structure

40
 Linked Representation
 Graph is represented using adjacency list
 For each node, it’s neighbor’s information is
maintained in a linked list fashion

 Adjacency list stores information about only those

edges that exist

 The linked representation contains two lists:

 A node list
 An edge list

41
Example

42
 Linked Representation
 Node List / Vertex List

Name or Key Pointer to Pointer to Other information (if any)

value of the the next the first e.g. indegree, outdegree,
node node in list element in status etc.
NODE list EDGE

 Edge List

Pointer to To link the

the location edges with
in the list the same
NODE of the initial node
destination 43
44
Linked Representation

45
 Adjacency Multilists
 Adjacency lists maintained as multilists

 Multilists  lists in which nodes may be

shared among several lists

 Exactly one node for each edge but this

node will be in two lists

46
 Adjacency Multilists

where
m is a boolean mark field
(may be used to check whether the edge has been
examined)

47
Adjacency Multilists - Example

48
 Graph ADT
typedef struct vertex
{
struct vertex* pNextVertex;
void* dataPtr;
struct edge* pEdge;
}VERTEX;

typedef struct edge

{
struct vertex* destination;
struct edge* pNextEdge;
}EDGE;

49
 Traversing a Graph
 Breadth First Traversal
 Uses a queue as an auxiliary structure to hold
nodes / vertices for future processing

 Depth First Traversal

 Uses a stack as an auxiliary structure to hold
nodes / vertices for future processing

50
 Algorithms
 During execution of algorithm each node N
of G will be in one of three states

 STATUS = 1 (Ready State) – Initial state of N

 STATUS = 2 (Waiting State) – N is on the
queue or stack, waiting to be processed
 STATUS = 3 (Processed State) – N has been
processed

51
 Breadth-first Traversal
 General Idea
 Process starting node say N
 Process all neighbors of N
 Process all neighbors of neighbors of N and so
on…
 No node should be processed more than once
 Use of queue to hold nodes that are waiting to
be processed

52
 Breadth First Traversal -
Algorithm
Algorithm BreadthFirstG (graph)
1. Initialize all nodes to the ready state (STATUS=1)
2. Put the starting node N in queue (enqueue N) and
change its status to the waiting state (STATUS=2)
3. loop(queue is not empty)
4. a. Remove the front node N of queue (dequeue N)
b. Process N and change the status of N to the
processed state (STATUS=3)
5. a. Add to the rear of queue all the neighbors of N
that are in ready state (STATUS=1)
b. Change their status to the waiting state
(STATUS=2)
6. end loop

end BreadthFirstG 53
Breadth First Traversal - Example

54
 Depth-first Traversal
 General Idea
 Examine starting node say N
 Along path P, Examine a neighbor of N, then a
neighbor of neighbor of N and so on…
 After coming to the “dead end” i.e. end of path
P, backtrack on P until we can continue along
another path P’ and so on…
 Use of stack to hold nodes that are waiting to
be processed

55
 Depth First Traversal - Algorithm
Algorithm DepthFirstG (graph)
1. Initialize all nodes to the ready state (STATUS=1)
2. Push the starting node N onto stack and change to the
waiting state (STATUS=2)
3. loop (stack is not empty)
4. a. Pop the top node N of stack
b. Process N and change the status of N to the
processed state (STATUS=3)
5. a. Push onto stack all the neighbors of N that are still
in the ready state (STATUS=1)
b. Change their status to the waiting state
(STATUS=2)
6. end loop

end DepthFirstG 56
Example

57
 Minimum Spanning Tree
 Spanning Tree
 A tree that contains all of the vertices in the
graph

 Minimum Spanning Tree

 A spanning tree in which the total weight of
the edges is guaranteed to be the minimum of
all possible trees in the graph

58
MST - Example

59
 Exercise - 1
 For the diagraph given below obtain
 The adjacency matrix
 The in-degree and out-degree of each vertex
 The adjacency-list representation
 The adjacency-multilist representation (assuming
undirected graph)

60
 Exercise - 2
 Is the graph strongly connected?
 List all the simple paths.
 Obtain the adjacency-matrix,
adjacency-list and
adjacency-multilist
representations
of the graph
 Obtain path matrix for
the graph

64
 Exercise - 3
 Apply depth first and breadth first traversal
to the complete graph of four vertices A, B,
C, D. List the vertices in the order that they
are visited.

65
 Exercise - 4
 Suppose a weighted graph G is maintained
in memory by a node array DATA and a
weight matrix W as below. Draw a picture
of G.

 DATA: X, Y, S, T

66
WCE - SY (CSE) – 1CI 202: Data Structures
Course Teacher: Dr. Smriti H. Bhandari