Graphs

Graphs
 Introduction to Graphs

How to describe a complex problem?

Problem Use Example

Simple/ Easy Some data Quadratic equation/ Student management/…
or a list of
items
Complex/ Difficult A graph • Transportation road map/road network?
(draw) Relations of airports
• Drainage net
• Country’s national electricity grid
• Computer networks

Graphs are tools for presenting a set of objects in

which each object can have relations to others.

Graphs 2
 Introduction to Graphs…
Examples:

(From Google map)

Approach 1: - Object: a device
- Object: a territory - Relation: connection
- Relation: Adjacency of 2 between two devices
territories - Object: an airport
Approach 2: - Relation: connection
- Object: crossroad between two airport
- Relation: road connecting
2 crossroads

Graphs 3
 Learning Outcomes

LO5.1 Demonstrate graphs and their applications.

Define an undirected graph and directed graph.
LO5.2 Explain basic notations about a graph: vertex,
edge, adjacent vertices, incident edge,...
LO5.3 Know how to represent a graph using matrix and
list
LO5.4 Able to apply BFS and DFS to traverse a graph
LO5.5 Write programs to implement Graph with BFS
and DFS traversals using Java
LO5.6 Explain the shortest path problem and Dijkstra’
algorithm.

Graphs 4
 Learning Outcomes

LO5.7 Write program in Java to implement Dijkstra’s

shortest path algorithm.
LO5.8 Define the minimum spanning tree and using the
Kruskal’s algorithm to find it.
LO5.9 Define Euler cycle/path and the necessary and
sufficient conditions for their existence.
LO5.10 Explain the algorithm to find Euler path / circuit
using Stack and implement it in Java language.
LO5.11 Define Halmilton cycle/path and able to
determine them using backtracking algorithm.
LO5.12 Demonstrate the graph coloring and can apply
Sequential coloring algorithm to color a graph.
Graphs 5
 Contents
1- Definitions
2- Graph Representation
Basic problems on a graph
3- Graph Traversals
4- Cycle Detection
5- Shortest Paths
6- Spanning Trees – cây phủ
7- Euler Cycles/Paths
8- Halmilton Cycles/Paths
9- Graph coloring
Codes:
- Basic codes: 2
- Comprehensive codes: 2

Graphs 6
 1- Definitions

• A graph (G) is a collection of vertices (or nodes,

the V set) and the connections (called as edges,
the E set between them. G = { V, E }
• A vertex is described as a circle.
• An edge is described as a link line between two vertices.
• Vertex label is information of a vertex which helps to
differentiate a vertex with others. In a drawing, the label of a
vertex is described using vertex’s key (character/integer).

vertex
vertex’s
label

edge

Graph is applied to manage a group of objects in which one object can

link to others.
Graphs 7
 1- Definitions…
Some Real Graphs:

Approach 1: Managing every thing

- Vertex: a territory.
- Edge: Adjacency of 2 territories

Approach 2: Managing roads only

- Vertex: a crossroad.
- Edge: road connecting 2 crossroads
Graphs 8
 1- Definitions…

• Graph Classification:
– Based on direction of edges:
– Based on number of edges connection
two vertices.
– Based on whether or not edges are
evaluated. An edge may have proper data
(a number).
 Mix graph: a graph having multiple
properties.

Graphs 9
 1- Definitions…

• Classification based on edge’s direction:

– Undirected graph / directed graph (digraph)
 In a directed graph, each edge has it’s own
direction (orientation)

Graphs 10
 1- Definitions…
• Classification based on number of
edges between vertices
– Simple graph: two distinct vertices in un-directed
graph can be join atmost ONE edge
– Multigraph is a graph in which two vertices can be
joined by multiple edges
– Pseudograph is a multigraph which allows for loops at
one vertex.

Graphs 11
 1- Definitions…

• Classification based on whether or not

edges are evaluated
– Weighted graph if each edge has an
assigned number (property of each edge)
– Un-weighted graph

Graphs 12
 1- Definitions…

Degree of a vertex – Bậc của đỉnh

• In undirected graphs: Degree of a
vertex, (deg(v), is number of edges
connecting with it.
• In directed graph:
• In-degree (bậc vào), deg+(v), is
number of edges connected to a
vertex.
• Out-degree, deg-(v), is number of
edges connected from a vertex.
• deg(v) = deg+(v) + deg-(v)

Graphs 13
 1- Definitions…
Path and circuit/circle
• A path is a group of consecutive edges.
• A circuit or circle is a closed path.

Graphs 14
 1- Definitions…
Connectivity:
A graph is connected if and only if there
exist a path between two arbitrary different
vertices.
- If a graph is not connected, it is a combination
of at least two connection components.

Connected graph This graph includes 3 Connected graph Weak connected

connected components (Strong connected)

A directed graph is weak connected if it’s associated undirected sub-graph is connected

Graphs 15
 1- Definitions…
Connectivity…
- Articulation point/cut vertex(điểm khớp/đỉnh cắt)
- Cut edge (cạnh cắt) / Bridge (cầu)
A special vetex or edge in a graph having
special property that when it is removed, it
causes the graph unconnected.

Graphs 16
 2- Graph Representation

2 ways to present data of a graph:

• Using matrix:
– adjacency matrix(ma trận đỉnh kề) or,
– incidence matrix ( ma trận cạnh nối)
It requires huge memory block
• Using adjacency list (danh sách kề)
– Linked data structure is used to save memory.

Graphs 17
 2- Graph Representation…
Representing a graph using a matrix:

Edge label = association of

labels of two related vertices
Vertices are sorted Edges are
based on their labels sorted

(a) an adjacency matrix (b) an incidence matrix

Graphs 18
 2- Graph Representation…
Some graphs using adjacency matrices:

Graphs 19
 2- Graph Representation…
Representing a graph using a matrix:

Evaluating on memory efficiency

Situation: Road map of a city has 10000 crossroads,
and at most 7 roads connecting each 2 distinct
crossroads.

Using adjacency matrix:

Matrix size: 10000 x 10000.
In each row, there is at most 7 non-zero numbers.
 Efficiency: (7*10000)/(10000 x 10000)  7/10000

Using incident matrix: Do yourself

Graphs 20
 2- Graph Representation…
Representing a graph using an adjacency list:

Adjacency
list of a
vertex

Vertex list

All lists are

sorted.

Graphs 21
 3- Graph Traversals

Graph traversal:
• An operation which will visit all vertices, each
vertex will be done only one time  O(n).
• Traversals on a graph are similar as those are
introduced in the chapter Trees.
– Breadth-First Traversal (Breadth-First Search- BFS)
– Depth- First Traversal (Depth-First Search- DFS)

Graphs 22
 3- Graph Traversals…

Breadth- First Search: Example

1 6 7 9
2

3
4 8

Figure 8-5 An example of application of the breadthFirstSearch()

algorithm to a graph

aefgibchd

Graphs 23
 3- Graph Traversals…

num(v) is a positive
breadthFirstSearch() // using a queue integer which
describes the visited
for all vertices v num(v) = 0; // reset visited orders
order of the vertex v
order = 1; // order begin from 1
while there is a vertex v such that num(v) is 0 // Visiting all vertices
num(v) = order++;
enqueue (v)
while queue is not empty
v=dequeue();
for all vertices u adjacent to v // considering all adjacent vertices of v
if num(u) is 0 {
num(u) = order++;
enqueue(u);
}

Graphs 24
 3- Graph Traversals…
Breadth First Search: Walking through the algorithm:

Result Queue
(a,1)
(a,1) (e,2),(f,3),(g,4),(i,5)
a efgi (a,1),(e,2) (f,3),(g,4),(i,5)
b g
(a,1),(e,2),(f3) (i,5)
c h
d h (a,1),(e,2),(f3),(g,4), (i,5),(b,6)
e afi (a,1),(e,2),(f3),(g,4),(i,5) (b,6)
f aei
(a,1),(e,2),(f3),(g,4),(i,5),(b,6) (c,7)
g ab
h cd (a,1),(e,2),(f3),(g,4),(i,5),(b,6),(c,7) (h,8)
i aef
(a,1),(e,2),(f3),(g,4),(i,5),(b,6),(c,7),(h,8) (d,9)

(a,1),(e,2),(f3),(g,4),(i,5),(b,6),(c,7),(h,8),(d,9)
Graphs 25
 3- Graph Traversals…
Breadth First Search: Walking through the algorithm:
(For taking quiz)
Queue 
aefgi
aefgI
aefgi
aefgib
aefgib
aefgibc
Một đỉnh đã được đưa vào hàng đợi rồi thì
aefgibch
gạch ngang (đánh dấu đã xử lý) aefgibchd
aefgibchd

Graphs 26
 3- Graph Traversals…

Breadth First Search: Check the result.

The breadthFirstSearch() algorithm applied to a digraph. Result:

a, e, f, i, b, g, c, h, d

Graphs 27
 3- Graph Traversals…

Practice: Apply the BFS algorithm to the following

graphs

G1 G2

Result:
1, …….. 1, ……..

Graphs 28
 3- Graph Traversals…
Depth-First Search algorithm:
Each vertex v is visited and then each
unvisited vertex adjacent to v is visited.
1
6 7 9
2

3
5
8
4

Graphs 29
 3- Graph Traversals…
Depth-first search algorithm: each vertex v is visited
and then each unvisited vertex adjacent to v is
visited
DFS (Vertex u) // Recursive algorithm
num(u) = order++; // i: visiting order
for all vertices v adjacent to u
if (num(v) is 0) // v is not traversed yet
attach edge uv to edges; // update result
DFS(v); // recursive traversing down
depthFirstSearch()
for all vertices v num(v) = 0;
edges = null; // Initiate the traverse result
order = 1;
while there is a vertex v such that num(v) is 0 DFS(v);
output edges;

Graphs 30
 3- Graph Traversals…
An example of application of the depthFirstSearch()

Graphs 31
 3- Graph Traversals…

• Depth First Search guarantees generating a tree

(or a forest, a set of trees) that includes or spans
over all vertices of the original graph.

The depthFirstSearch() algorithm applied to a digraph

Graphs 32
 3-Graph Traversals…

Practice: Apply the DFS algorithm to the following

graphs:

G1 G2

Graphs 33
 4- Cycle Detection

• Depth-First Search  Cycle Detection

DFS (Vertex v)
digraphCycleDetectionDFS (Vertex v)
num(v) = i++;
num(v) = i++;
for all vertices u adjacent to v for all vertices u adjacent to v
if (num(u) is 0) if (num(u) is 0)
modify pred (u)= v;
attach edge uv to edges;
digraphCycleDetectionDFS(u);
DFS(u); else if num(u) is not 
cycleDetectionDFS (Vertex v) pred (u) = v;
cycle detected;
num(v) = i++;
num(v) = ;
for all vertices u adjacent to v
if (num(u) is 0) // u is not visited  continue traversing
attach edge uv to edges;
{ae, ef, fa}
O(n2) cycleDetectionDFS(u);
else if edge(uv) is not in edges, cycle detected // u is re-visited  cycle

Graphs 34
 Lab 1- Demo 1: Traversing a Graph
(Using Adjacency Matrix)
Objectives:
- Traverse a
simple
undirected
matrix- Graph_Matrix: class for a graph.
MyQueue: Class for a queue
based graph TestTraverser: Tester
- Output
traversal
results to Output files
monitor and
files. Input file

Graphs 35
 Demo 1: Traversing a Graph

Input:

Graphs 36
 Demo 1: Traversing a Graph
MyQueue.java

Graphs 37
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 38
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 39
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 40
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 41
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 42
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 43
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 44
 Demo 1: Traversing a Graph

Graph_Matrix.java

Graphs 45
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 46
 Demo 1: Traversing a Graph
Graph_Matrix.java

Graphs 47
 Demo 1: Traversing a Graph
TestTraverser.java

Graphs 48
 Demo 1: Traversing a Graph
TestTraverser.java

Graphs 49
 Demo 1: Traversing a Graph
TestTraverser.java

Graphs 50
 Demo 1: Traversing a Graph
TestTraverser.java

Graphs 51
 5- Shortest Paths

• Weight of an edge:
– A real number which depends on specific
problem.
– It is positive but it may be negative in some
cases.
• Road map: road length (meter)
Shortest
• Aero map: fly length/ cost ($) Paths are
• Computer network: bandwidth (bit/sec) basic
• Drainage net: water flow (m /s)
3
requirement
• Electricity grid: load power (watt) in Graph
• .... theory
• It may be a multiple-variable function

Graphs 52
 5- Shortest Paths…
• Label: considered values are applied to
each vertex. In this Dijkstra, introduced,
Label(v) = {cost(v), predecessor(v)}
• To get shortest path between two vertices, each
vertex must be chosen to scan for evaluating. In
each pass, only one vertex is chosen to scan.
• Classification:
– Label-setting methods: In each pass through the
vertices still to be processed, one vertex is set to a
value that remains unchanged to the end of the
execution, the Dijkstra algorithm is an example.
– Label-correcting methods: Allow for the changing of
any label during application of the method

Graphs 53
 5- Shortest Paths…
Algorithms:
Main algorithm:
• Dijkstra — weight of edge >=0
Algorithms should be referenced:
• Bellman-Ford — weight of edge may be negative.
• A*: Using heuristics to improve performance
• Floyd-Warshall — Determine shortest path for all pair of
vertices.
• Johnson — Determine shortest path for all pair of
vertices. It can be more efficient than Floyd-Warshall on
a sparse graph.
Attention: Each algorithm has it’s own purpose and a specific graph type.

Graphs 54
 5- Shortest Paths…
GenericShortestPathAlgorithm (weighted simple digraph,
vertex first) Dijkstra
for all vertices v cost(v) =  toBeChecked= all vertices
cost(first)=0;
initialize toBeChecked; // set of vertices will be checked
while (toBeChecked is not empty)
v= a vertex in toBeChecked; // depending on specific algorithm
remove v from toBeChecked;
Ford
for all vertices u adjacent to v Examine edges instead of vertices
if cost(u) > cost(v) + weight(v,u))
cost(u) = cost(v) + weight(v,u));
O(|V|2) predecessor(u)= v;
Add u to toBeChecked if it is not there;

Graphs 55
 5- Shortest Paths…
DJ ShortestPathAlgorithm (weighted simple digraph, vertex first)
for all vertices v {
marked(v) = false; cost(v) = ; predecessor (v) = null;
}
cost(first) = 0;
v = first;
while (v != null) {
marked (v) = true;
for all vertices u adjacent to v {
if (marked (u)=false AND (cost(u) > cost(v) + weight(v,u))) {
cost(u) = cost(v) + weight(v, u));
predecessor(u)= v;
}
}
v= the minimum cost vertex in non-marked vertices;
}
Graphs 56
 5- Shortest Paths…

Finding shortest paths from the vertex d to others

V = infinity

4,d- 9,f- 11,f-

5,a- 12,j

8,e-
0,null-

1,d- 9,f- 11,i-

An execution of DijkstraAlgorithm
(1) Độ dài đường đi ngắn nhất từ d sang i =?
(2) Đường đi ngắn nhất từ d sang i là đường nào?
(3) Sau bước lặp thứ 3 của thuật toán DJ, độ dài đường đi ngắn nhất từ d đến i =?  10
Graphs 57
 5- Shortest Paths…

Exercise: Use Dijkstra algorithm to find the shortest

path from the vertex 5 to the vertex 4

Graphs 58
 Lab 2- Demo 2: SP Dijkstra Algorithm

Dijkstra
Algorithm Impl.

Input
output file

Graphs 59
 Demo 2: SP Dijkstra Algorithm
DJ_Finder.java

Graphs 60
 Demo 2: SP Dijkstra Algorithm
DJ_Finder.java

Graphs 61
 Demo 2: SP Dijkstra Algorithm
DJ_Finder.java

Graphs 62
 Demo 2: SP Dijkstra Algorithm
DJ_Finder.java

Graphs 63
 Demo 2: SP Dijkstra Algorithm
DJ_Finder.java

Graphs 64
 Demo 2: SP Dijkstra Algorithm
DJ_Test.java

Graphs 65
 Demo 2: SP Dijkstra Algorithm

Graphs 66
 6- Spanning Trees- Cây phủ
Subgraph: G’ (V’, E’) is called as a subgraph of the
graph G(V, E) if and only if V’ is a subset of V and E’ is
a subset of E.
Spanning Tree: G’ (V’, E’) is called as a spanning tree
of subgraph of the graph G(V, E) if and only if V’=V and
E’ is a subset of E and G’ is a tree (connected, having
no cycle).

A graph
may have
some
spanning
trees.

A graph representing (a) the airline connections between seven

cities and (b–d) three possible sets of connections
Graphs 67
 6- Spanning Trees…

- A graph may have some spanning trees

- In weighted graph, minimum spanning tree is a
spanning tree which sum of edge weights is
minimum.
How to construct minimum spanning trees?
Algorithms will be introduced:
- Krushkal (written in learning outcome)
- Dijkstra ( Read by yourself)
- Prim-Jarnik ( Read by yourself)

Graphs 68
 6- Spanning Trees…

• Minimum spanning tree- Kruskal algorithm:

– Input: weighted connected, undirected graph
– Output: minimum spanning tree
• tree = null;
• edges = sequence of all edges of graph sorted by
weight;
• for (i=1; i<= |E| and |tree| < |V| -1; i++)
– if ei from edges does not form a cycle with edges in tree
add ei to tree.
Edges having small weight are choosen with high priority

Graphs 69
 6- Spanning Trees…
• Minimum spanning tree-
Kruskal algorithm:
• Lấy cạnh có trọng số bé mà
không tạo thành vòng.

Graphs 70
 6- Spanning Trees…

• Read by yourself
• Minimum spanning tree- Dijkstra algorithm:
– Input: weighted connected, undirected graph
– Output: minimum spanning tree
• tree = null;
• edges = an unsorted sequence of all edges of graph ;
• for j=1 to |E|
- add ej to tree;
- if there is a cycle in tree, remove an edge with
maximum weight from this only cycle;

Graphs 71
 6- Spanning Trees…
• Read by yourself
• Minimum spanning tree- Dijkstra
algorithm:

Edges are
choosen in
natural order
of edge
abels

Graphs 72
 6- Spanning Trees…

• Read by yourself
• Minimum spanning tree- Prim-Jarnik algorithm:

Step 1: Initialize a tree with a single vertex, chosen

arbitrarily from the graph.
Step 2: Grow the tree by one edge of the edges that
connect the tree to vertices not yet in the tree,
find the minimum-weight edge, and transfer it to
the tree.
Step 3: Repeat step 2 (until all vertices are in the
tree).

Graphs 73
 6- Spanning Trees…
74

Read by yourself
Prim-Jarnik Algorithm demo
A spanning tree of
graph (a) found
with Prim’s
algorithm starting
with the vertex 0.
The vertices are
selected in the
following order:
0, 1, 7, 6, 5, 2, 8,
3, 4

74/30
Graphs
 7- Euler Cycles

Euler path: a path traversing all the edges of the graph

exactly once
Euler cycle: a cycle traversing all the edges of the
graph exactly once

Has Euler cycle No Euler cycle

no Euler path

Has Euler path,

but no Euler cycle

Graphs 75
 7- Euler Cycles…

C
Is it possible to start at some
A location, travel across all the
D bridges without crossing any
bridge twice, and return to
B the same starting point?
C
Kneiphof island on Pregel river
and 7 bridges built in 18-th
century in Königsberg town A D
(Kaliningrad).

Rephrasing in terms of Euler cycles: B

Graphs 76
 7- Euler Cycles…
Necessary and sufficient conditions for Euler cycles/ Euler path

Undirected graph
Theorem 1: A connected multigraph has an Euler cycle if
and only if each of its vertices has even degree.
Theorem 2: A connected multigraph has an Euler path
but Euler cycle if and only if it has exactly two odd degree
vertices.

Which of
graphs have
Euler cyles/
Euler path?

Graphs 77
 7- Euler Cycles…

Directed graph

- A connected directed multigraph has an Euler cycle if and only if

deg+(v) = deg-(v) for all vertex v.
- A connected directed multigraph has an Euler path but Euler
cycle if and only if it has exactly two vertices u and v in which one
vertex having deg+(u) – deg-(u) = +1 and deg+(v) – deg-(v) = -1.

1,2

Which of
graphs have
Euler cyles/
Euler path?

Graphs 78
 7- Euler Cycles…
Algorithm Euler for a graph having Euler cycle.
Main Idea: Splicing all sub-cycles

(1) Copy G cycle

(2) Contruct a cycle
(1,2,3) then move
others to H
(3) Loop until H
having no edge:
From a common
vertex (ex. 3) of
cycle and H,
construct a new
cycle then splice
this to cycle.

Graphs 79
 7- Euler Cycles…

Algorithm Euler(G)
Input: Connected graph G with all vertices having even degrees
Output: Euler cycle
Construct a cycle in G
Remove all the edges of cycle from G to get subgraph H
while H has edges
find a non-isolated vertex v that is both in cycle and in H
//the existence of such a vertex is guaranteed by G’s connectivity
construct subcycle in H
splice subcycle into cycle at v
remove all the edges of subcycle from H
return cycle

Graphs 80
 7- Euler Cycles…
Algorithm for finding an Euler cycle from
the vertex X using stack
Algorithm Euler(G, start) Chuyển các
//Input: Connected graph G with all vertices having even degrees chu trình
//Output: Euler cycle
con vào
declare a stack S of vertices which will be examined stack, giảm
declare empty array E (which will contain Euler cycle) bậc mỗi
push the vertex start to S
while(S is not empty) { đỉnh của
u= the vertex on the top of the stack S các chu
if (u is isolated) then remove it from the stack and put it to E trình này 2
else
v= an arbitrary vertex which is adjacent to u, đơn vị cho
push v to S đến khi bậc
Remove edges (u, v) and (v,u) from the graph của đỉnh
}
Vertices in E form a Euler cycle of the graph bằng 0.

Graphs 81
 Lab 3- Demo 3: Euler Cycle
Finding an Euler cycle using stack

A C E G

B D F H

Implementations

Input/
output files

Graphs 82
 Euler_Graph.java Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 83
 Euler_Graph.java Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 84
 Euler_Graph.java Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 85
 Euler_Graph.java Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 86
 Euler_Graph.java Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 87
 Euler_Graph_Test.java
Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 88
 Euler_Graph_Test.java
Demo 3: Euler Cycle
Finding an Euler cycle using stack

Graphs 89
 8- Hamilton Cycles
Hamilton cycle: visits every vertex of the graph
exactly once before returning, as the last step, to the
starting vertex.

Examples:

Hamilton path: visits every vertex of the graph exactly once.

Graphs 90
 8- Hamilton Cycles…
Finding Hamilton’s cycles
using Backtracking
Choose 1 in 6 vertices
(Exhausted Searching – vét
Choose 1 in 5 vertices
cạn- dò tìm cho đến cùng)
Choose 1 in 4 vertices
Choose 1 in 3 vertices
Choose 1 in 2 vertices
Choose 1 in 1 vertices

7 vertices:
6! options
A C B E G F D A

A A A A A A
B B B B B B O(n!)  AI is needed
C C C C C C
D D D D D D
E E E E E E
F F F F F F
G G G G G G

Graphs 91
 8- Hamilton Cycles…
A graph can have some Hamilton cycles.
List all Hamilton’s cycles using Backtracking

The given graph has four Hamilton circles.

Graphs 92
 8- Hamilton Cycles…
Finding Hamilton’s cycles using Backtracking
(Exhausted Searching – vét cạn- dò tìm cho đến
cùng)
Given the graph G = (V,E) and X is a vertex of G. Suppose
there exists at least one Hamilton Cycle for the graph. The
following is a backtracking algorithm for finding one Hamilton
cycle from the vertex X:
declare an empty array H (which will contain Hamilton cycle)
(1) Put the vertex X to H
(2) Check if H is a Hamilton cycle then stop, else go to (3)
(3) Consider the last vertex Y in H, if there is/are vertex(es)
adjacent to Y, select an adjacent vertex Z and put it to H. If
there no adjacent vertex, remove Y from H and denote it as a
bad selection (so you do not select it in the same way again).
Go to (2). O(n!)  AI is needed
Graphs 93
 9- Graph Coloring Problem

• In graph theory, graph coloring is a way of coloring the vertices of

a graph such that no two adjacent vertices share the same color.
• The chromatic number (số mầu) of a
graph is the minimum number of colors
one can use to color the vertices of the
graph so that no two adjacent vertices are
the same color.
• If the chromatic number of the graph G
is denoted by χ(G). A graph for which k =
χ(G) is called k-colorable. For a
complete
graph χ(Kn)=n, χ(C2n)=2, χ(C2n+1)=3; and for bipartite graph χ(G)≤ 2.
• Determining a chromatic number of a graph G is an NP-complete
problem.
Graphs 94
 9- Graph
Coloring
Problem…
Sequential coloring:
(a) A graph used for coloring;
(b) Colors assigned to
vertices with
the sequential coloring
algorithm that orders
vertices by index number
Changing
(c) Vertices are put in the order of
largest first sequence; vertices
(d) Graph coloring obtained
with the Brélaz algorithm

Graphs 95
 9- Graph Coloring Problem…
• Sequential coloring establishes the sequence of
vertices and a sequence of colors before coloring
them, and then color the next vertex with the
lowest number possible
sequentialColoringAlgorithm(graph = (V, E))
put vertices in a certain order vP1, vP2, . vpi . . , vPv ;
put colors in a certain order c1, c2, . . . , ck;
for i = 1 to |V|
j=the smallest index of color that does not appear in any neighbor of vPi;
color(vPi) = cj;

Complexity: O(|V|2)

Graphs 96
 Summary

 LO5.1 Demonstrate graphs and their

applications. Define an undirected graph and
directed graph.
 LO5.2 Explain basic notions about a graph:
vertex, edge, adjacent vertices, incident
edge,...
 LO5.3 Know how to represent a graph using
matrix and list
 LO5.4 Able to apply BFS and DFS to traverse a
graph
 LO5.5 Write programs to implement Graph
with BFS and DFS traversals using Java
Graphs 97
 Summary

LO5.6 Explain the shortest path problem and

 Dijkstra’ algorithm.
LO5.7 Write program in Java to implement
 Dijkstra’s shortest path algorithm.
LO5.8 Define the minimum spanning tree and
 using the Kruskal’s algorithm to find it.
LO5.9 Define Euler cycle/path and the
 necessary and sufficient conditions for
their existence.
LO5.10 Explain the algorithm to find Euler
 path / circuit using Stack and implement it
in Java language.
Graphs 98
 Summary

 LO5.11 Define Halmilton cycle/path and

able to determine them using
backtracking algorithm.
LO5.12 Demonstrate the graph coloring

and can apply Sequential coloring
algorithm to color a graph.

Graphs 99