Posts and Telecommunications Institute of Technology

Faculty of Information Technology 1

Discrete Mathematics 2

Graph Representation in Computers

Ngo Xuan Bach

 Contents
 Graph representation using adjacency matrix
 Graph representation using incidence matrix
 Graph representation using edge list
 Graph representation using adjacency list

2 http://www.ptit.edu.vn
 Adjacency Matrix of Undirected Graph
 Consider undirected graph 𝐺 =< 𝑉, 𝐸 >, 𝑉 = {1,2, … , 𝑛} is
the set of vertices, 𝐸 = {𝑒1 , 𝑒2 , . . . , 𝑒𝑚 } is a set of edges.
The adjacency matrix of 𝐺 is a square 𝑛 × 𝑛 matrix,
whose elements equal to 0 or 1:
o 𝐴 = {𝑎𝑖𝑗 : 𝑎𝑖𝑗 = 1 𝑖𝑓 (𝑖, 𝑗) ∈ 𝐸, 𝑎𝑖𝑗 = 0 𝑖𝑓 (𝑖, 𝑗) ∉ 𝐸; 𝑖, 𝑗 = 1, 2, . . . , 𝑛}.

3 http://www.ptit.edu.vn
 Adjacency Matrix of Undirected Graph
 Consider undirected graph 𝐺 =< 𝑉, 𝐸 >, 𝑉 = {1,2, … , 𝑛} is
the set of vertices, 𝐸 = {𝑒1 , 𝑒2 , . . . , 𝑒𝑚 } is a set of edges.
The adjacency matrix of 𝐺 is a square 𝑛 × 𝑛 matrix,
whose elements equal to 0 or 1:
o 𝐴 = {𝑎𝑖𝑗 : 𝑎𝑖𝑗 = 1 𝑖𝑓 (𝑖, 𝑗) ∈ 𝐸, 𝑎𝑖𝑗 = 0 𝑖𝑓 (𝑖, 𝑗) ∉ 𝐸; 𝑖, 𝑗 = 1, 2, . . . , 𝑛}.

0 1

4 http://www.ptit.edu.vn
 Properties of Adjacency Matrix
of Undirected Graph
 Symmetry: 𝑎𝑖𝑗 = 𝑎𝑗𝑖
 Sum of all the elements equals to two times of the number of
edges
𝑛 𝑛
o 𝑖=1 𝑗=1 𝑎𝑖𝑗 = 2|𝐸|
 Sum of all the elements in row 𝑢 equals to the degree of
vertex 𝑢
𝑛
o 𝑗=1 𝑎𝑢𝑗 = deg(𝑢)
 Sum of all the elements in column 𝑢 equals to the degree of
vertex 𝑢
𝑛
o 𝑖=1 𝑎𝑖𝑢 = deg(𝑢)
𝑝
 Let 𝑎𝑖𝑗 (𝑖, 𝑗 = 1,2, … , 𝑛) be the elements of matrix 𝐴𝑝 =
𝑝
𝐴. 𝐴. . . 𝐴 (𝑝 times), then 𝑎𝑖𝑗 is the number of distinct paths
from vertex 𝑖 to vertex 𝑗 through 𝑝 − 1 intermediate vertices

5 http://www.ptit.edu.vn
 Adjacency Matrix of Directed Graph
 Similar to the adjacency matrix of undirected graph
o There exists a direction on each arrow
o Not symmetric

6 http://www.ptit.edu.vn
 Adjacency Matrix of Directed Graph
 Similar to the adjacency matrix of undirected graph
o There exists a direction on each arrow
o Not symmetric

7 http://www.ptit.edu.vn
 Properties of Adjacency Matrix
of Directed Graph
 Sum of all the elements equals to the number of edges
𝑛 𝑛
o 𝑖=1 𝑗=1 𝑎𝑖𝑗 = |𝐸|
 Sum of all the elements in row 𝑢 equals to the outdegree
of vertex 𝑢
𝑛
o 𝑗=1 𝑎𝑢𝑗 = 𝑑𝑒𝑔+ (𝑢)
 Sum of all the elements in column 𝑢 equals to the
indegree of vertex 𝑢
𝑛
o 𝑖=1 𝑎𝑖𝑢 = 𝑑𝑒𝑔− (𝑢)
𝑝
 Let 𝑎𝑖𝑗 (𝑖, 𝑗 = 1,2, … , 𝑛) be the elements of matrix
𝑝
𝐴𝑝 = 𝐴. 𝐴. . . 𝐴 (𝑝 times), then 𝑎𝑖𝑗 is the number of distinct
paths from vertex 𝑖 to vertex 𝑗 through 𝑝 − 1
intermediate vertices

8 http://www.ptit.edu.vn
 Weighted (Adjacency) Matrix
 A real value 𝑐(𝑒) = 𝑐(𝑢, 𝑣) is assigned to each edge
𝑒 = (𝑢, 𝑣), called the weight of edge e
o In this case the graph is called weighted graph
o Weighted matrix 𝑐 = 𝑐[𝑖, 𝑗], 𝑐[𝑖, 𝑗] = 𝑐(𝑖, 𝑗) if (𝑖, 𝑗) ∈ 𝐸, 𝑐[𝑖, 𝑗] = 𝜃 if
(𝑖, 𝑗) ∉ 𝐸. 𝜃 can received values 0, ∞, −∞ depending on the
problem

9 http://www.ptit.edu.vn
 Advantages & Disadvantages
of Adjacency Matrix
 Advantages
o Simple, easy to be implemented
 Using a two dimensional array to represent
o Easy to check whether two vertices 𝑢, 𝑣 are adjacent or not
 Only one comparison operation (𝑎[𝑢][𝑣] ≠ 0?)

 Disadvantages
o Waste of memory: we need 𝑛2 memory units despite the number
of edges
o Cannot represent a graph with a large number of vertices
o To find adjacency vertices of vertex 𝑢 we need 𝑛 comparison
operations despite the degree of 𝑢

10 http://www.ptit.edu.vn
 Adjacency Matrix Storage
 The first line is the number of vertices
 The next 𝑛 lines store the adjacency matrix
o Two matrix elements are separated by one or more spaces

11 http://www.ptit.edu.vn
 Contents
 Graph representation using adjacency matrix
 Graph representation using incidence matrix
 Graph representation using edge list
 Graph representation using adjacency list

12 http://www.ptit.edu.vn
 Incidence Matrix of Undirected Graph
 Consider undirected graph 𝐺 = (𝑉, 𝐸), 𝑉 = {1,2, … , 𝑛},
𝐸 = {𝑒1 , 𝑒2 , … , 𝑒𝑚 }. Vertex-edge incidence matrix of 𝐺 is a
𝑛 × 𝑚 matrix:
1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑎𝑛𝑑 𝑒𝑑𝑔𝑒 𝑗 𝑎𝑟𝑒 𝑖𝑛𝑐𝑖𝑑𝑒𝑛𝑡
𝑎𝑖𝑗 =
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

6
9
1
7
3 5
2 8
4

13 http://www.ptit.edu.vn
 Incidence Matrix of Undirected Graph
 Consider undirected graph 𝐺 = (𝑉, 𝐸), 𝑉 = {1,2, … , 𝑛},
𝐸 = {𝑒1 , 𝑒2 , … , 𝑒𝑚 }. Vertex-edge incidence matrix of 𝐺 is a
𝑛 × 𝑚 matrix:
1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑎𝑛𝑑 𝑒𝑑𝑔𝑒 𝑗 𝑎𝑟𝑒 𝑖𝑛𝑐𝑖𝑑𝑒𝑛𝑡
𝑎𝑖𝑗 =
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

6 1 2 3 4 5 6 7 8 9
9 1 1 1 0 0 0 0 0 0 0
1
7 2 1 0 1 0 1 1 0 0 0
3 5 3 0 1 1 1 0 0 0 0 0

2 8 4 0 0 0 1 1 0 1 1 0
5 0 0 0 0 0 1 1 0 1
4
6 0 0 0 0 0 0 0 1 1

14 http://www.ptit.edu.vn
 Incidence Matrix of Directed Graph
 Consider directed graph 𝐺 = (𝑉, 𝐸), 𝑉 = {1,2, … , 𝑛},
𝐸 = {𝑒1 , 𝑒2 , … , 𝑒𝑚 }. Vertex-arrow incidence matrix of 𝐺 is
a 𝑛 × 𝑚 matrix:
1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑖𝑠 𝑡ℎ𝑒 𝑠𝑡𝑎𝑟𝑡𝑖𝑛𝑔 𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑎𝑟𝑟𝑜𝑤 𝑒𝑗
𝑎𝑖𝑗 = −1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑖𝑠 𝑡ℎ𝑒 𝑒𝑛𝑑𝑖𝑛𝑔 𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑎𝑟𝑟𝑜𝑤 𝑒𝑗
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

6
1 9
3 7
5
2 8
4

15 http://www.ptit.edu.vn
 Incidence Matrix of Directed Graph
 Consider directed graph 𝐺 = (𝑉, 𝐸), 𝑉 = {1,2, … , 𝑛},
𝐸 = {𝑒1 , 𝑒2 , … , 𝑒𝑚 }. Vertex-arrow incidence matrix of 𝐺 is
a 𝑛 × 𝑚 matrix:
1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑖𝑠 𝑡ℎ𝑒 𝑠𝑡𝑎𝑟𝑡𝑖𝑛𝑔 𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑎𝑟𝑟𝑜𝑤 𝑒𝑗
𝑎𝑖𝑗 = −1, 𝑖𝑓 𝑣𝑒𝑟𝑡𝑒𝑥 𝑖 𝑖𝑠 𝑡ℎ𝑒 𝑒𝑛𝑑𝑖𝑛𝑔 𝑝𝑜𝑖𝑛𝑡 𝑜𝑓 𝑎𝑟𝑟𝑜𝑤 𝑒𝑗
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

6 1 2 3 4 5 6 7 8 9
1 1 -1 0 0 0 0 0 0 0
1 9
2 -1 0 1 0 1 1 0 0 0
3 7 3 0 1 -1 -1 0 0 0 0 0
5
4 0 0 0 1 -1 0 1 -1 0
2 8
5 0 0 0 0 0 -1 -1 0 1
4 6 0 0 0 0 0 0 0 1 -1

16 http://www.ptit.edu.vn
 Contents
 Graph representation using adjacency matrix
 Graph representation using incidence matrix
 Graph representation using edge list
 Graph representation using adjacency list

17 http://www.ptit.edu.vn
 Edge (Arrow) List
 In the case of sparse graphs (𝑚 ≤ 6𝑛), we usually use
edge (arrow) list
o We store all the edges (arrows) of the undirected graph (directed
graph). Each edge (arrow) 𝑒(𝑥, 𝑦) is represented by two variables
𝑠𝑡𝑎𝑟𝑡[𝑒], 𝑒𝑛𝑑[𝑒].
o Advantages: we need only 2𝑚 memory units to store the graph
o Disadvantages: to know which vertices are adjacent to a vertex
we need 𝑚 comparison operations (examine all 𝑚 edges)
o For weighted graph, we need more 𝑚 memory units to store
weights of edges

18 http://www.ptit.edu.vn
 Edge List of Undirected Graph
 We only list (𝑢, 𝑣), do not list (𝑣, 𝑢)
 Should list edges in the ascending order of the starting
points of edges
 The number of edges containing 𝑢 (left or right) is the
degree of vertex 𝑢
Starting point Ending point

19 http://www.ptit.edu.vn
 Arrow List of Directed Graph
 Each arrow is an ordered pair of vertices
o Starting point does not have to be less than ending point
 Number of arrows with 𝑢 on the left hand side is 𝑑𝑒𝑔+ (𝑢)
 Number of arrows with 𝑢 on the right hand side is
𝑑𝑒𝑔− (𝑢)
Starting point Ending point

20 http://www.ptit.edu.vn
 Edge List of Weighted Graph
 Add a new column for weights

Starting point Ending point Weight

21 http://www.ptit.edu.vn
 Advantages & Disadvantages of Edge List
 Advantages
o Save memory in the case of sparse graphs (𝑚 < 6𝑛)
o Convenient for algorithms considering only edges of graphs

 Disadvantages
o To now which vertices are adjacent to vertex 𝑢 we have to
examine all the edges of the graph
 High computational complexity

22 http://www.ptit.edu.vn
 Edge List Storage
 The first line stores 𝑁, 𝑀 , the number of vertices and the
number of edges of the graph
o Two numbers are separated by one or more spaces
 In the next 𝑀 lines, each line stores an edge of the graph
o Starting point and ending point are separated by one or more
spaces

23 http://www.ptit.edu.vn
 Data Structure for Edge List

//Definition of an edge
typedef struct {
int start;
int end;
int weight;
} Edge;

//Edge list is represented in array G

Edge G[MAX];

24 http://www.ptit.edu.vn
 Contents
 Graph representation using adjacency matrix
 Graph representation using incidence matrix
 Graph representation using edge list
 Graph representation using adjacency list

25 http://www.ptit.edu.vn
 Adjacency List
 For each vertex 𝑢 we store the list of its adjacent vertices
denoted by 𝐴𝑑𝑗(𝑢)
o 𝐴𝑑𝑗(𝑢) = { 𝑣 ∈ 𝑉: (𝑢, 𝑣) ∈ 𝐸}

Adj }
Adj
Adj
Adj
Adj
Adj

26 http://www.ptit.edu.vn
 Advantages & Disadvantages
of Adjacency List
 Advantages
o Easy to traverse all the vertices of an adjacency list
o Easy to traverse the edges of the graph in each adjacency list
o Optimal representation

 Disadvantages
o Difficult for people who are not good at programming

27 http://www.ptit.edu.vn
 Adjacency List using Array
 Array consists of 𝑛 segments
o The 𝑖 𝑡ℎ segment stores the adjacency list of the 𝑖 𝑡ℎ vertex
o To know where a segment starts or ends, we use another array
storing the starting and ending elements of the segment

Segment 1 Segment 2 Segment 3 Segment 4 Segment 5 Segment 6

Adj }
Adj
Adj
Adj
Adj
Adj

28 http://www.ptit.edu.vn
 Adjacency List using Linked List
 For each vertex 𝑢 ∈ 𝑉, represents the adjacency list of
the vertex using a linked list 𝐿𝑖𝑠𝑡(𝑢)

29 http://www.ptit.edu.vn
 Adjacency List Storage
 The first line stores the number of vertices
 Next 𝑁 lines store adjacency lists of vertices as follows:
o The first number is the ending position of the segment, other
numbers are list of vertices of the linked list
o Numbers are separated by one or more spaces

30 http://www.ptit.edu.vn
 Exercise 1
 In a meeting, some visitors make a handshake with some
others. Prove that the total number of handshakes of all
the visitors is even.

31 http://www.ptit.edu.vn
 Exercise 2
 What is the maximum number of edges that a simple
undirected graph with 𝑛 vertices can has?

32 http://www.ptit.edu.vn
 Exercise 3
 Represent the undirected graph below using:
1) Adjacency matrix
2) Edge list
3) Adjacency list

2 5

1 7
4

3 6

33 http://www.ptit.edu.vn
 Exercise 4
 Represent the directed graph below using:
1) Adjacency matrix
2) Edge list
3) Adjacency list

2 5

1 7
4

3 6

34 http://www.ptit.edu.vn
 Exercise 5
 Represent the weighted graph below using:
1) Weighted matrix
2) Edge list

3
2 5
2 8
6 1
1 2 5
4 7

2
4 7 5

3 4 6

35 http://www.ptit.edu.vn