Minimum Spanning Trees

In this section of notes you will learn two

algorithms for creating a connected graph
at minimum cost as well as a method for
ordering a graph

James Tam

Minimum Spanning Trees

Applies to weighted, undirected and connected graph
Create the minimum number of edges/arcs so that all
nodes/vertices are connected

James Tam

Example Of A Problem Dealing With Minimum

Spanning Trees

James Tam

Example Of A Problem Dealing With Minimum

Spanning Trees

James Tam

Example Of A Problem Dealing With Minimum

Spanning Trees

$$$

James Tam

Example Of A Problem Dealing With Minimum

Spanning Trees

James Tam

Algorithms For Determining The Minimum

Spanning Tree
Prims Algorithm
Kruskals Algorithm

James Tam

Prims Algorithm
primsAlgorithm (Graph g, Tree t, Node start)
{
PriorityQueue nodesLeft = new PriorityQueue();
Node temp;
Node neighbor;
for (int i = 1; i <= g.noNodes (); i++)
{
g[i].setWeight = ;
g[i].setParent (null);
nodesLeft.add (g[i]);
}
start.setWeight (0);
Pseudo code is based roughly on the algorithm provided by Matthew A. Becker
http://www.andrew.cmu.edu/user/mbecker/

James Tam

Prims Algorithm (2)

while (nodesLeft.isEmpty() == false)
{
temp = nodesLeft.dequeueMinWeightNode ();
t.add (temp);
for each node neighbor which is adjacent to temp and is in nodesLeft
{
if (weight (temp, neighbor) < neighbor.getWeight())
{
neighbor.setParent (temp);
neighbor.setWeight (weight (temp, neighbor));
}
}
}
}
Pseudo code based roughly on the algorithm provided by Matthew A. Becker
http://www.andrew.cmu.edu/user/mbecker/

James Tam

Example: Finding The Minimum Spanning Tree In

A Graph (Prims Algorithm)
6

e
8

f
2

James Tam

Initialized Values For The Graph And Queue

Graph

Tree
6

e
f

Queue
Weight

Node

A B C D E F G H I

Predecessor

0 0 0 0 0 0 0

0 0
James Tam

Mark A As Visited, Update Neighbors

Graph

Tree
6

e
f

Queue
Weight

0 6 4 2

Node

A B C D E F G H I

Predecessor

0 A 0 0 0 A 0

0 A
James Tam

Graph

Mark Node I As Visited And Include

The Edge (A,I)
Tree

e
f

Queue
Weight

0 6 4 2

Node

A B C D E F G H I

Predecessor

0 A 0 0 0 A 0

0 A
James Tam

Graph

Mark Node F As Visisted And Include

The Edge (A,F)
Tree

e
f

Queue
Weight

0 6 4 2

Node

A B C D E F G H I

Predecessor

0 A 0 0 0 A F

0 A
James Tam

Graph

Mark Node G As Visisted And Include

The Edge (F,G)
Tree

e
f

Queue
Weight

0 6 5 8 4 2

Node

A B C D E F G H I

Predecessor

0 A 0 G G A F

0 A
James Tam

Graph

Mark Node D As Visisted And Include

The Edge (D,G)
Tree

e
5

Queue
Weight

0 6 4 5 8 4 2

1 2

Node

A B C D E F G H I

Predecessor

0 A D G G A F

D A
James Tam

Graph

Mark Node H As Visisted And Include

The Edge (D,H)
Tree

e
5

d
g

Queue
Weight

0 6 4 5 8 4 2

1 2

Node

A B C D E F G H I

Predecessor

0 A D G G A F

D A
James Tam

Graph

Mark Node C As Visisted And Include

The Edge (C,D)
Tree

e
5

Queue
Weight

0 6 4 5 3 4 2

1 2

Node

A B C D E F G H I

Predecessor

0 A D G C A F

D A
James Tam

Graph

Mark Node E As Visisted And Include

The Edge (C,E)
Tree

d
f

Queue
Weight

0 6 4 5 3 4 2

1 2

Node

A B C D E F G H I

Predecessor

0 A D G C A F

D A
James Tam

Graph

Mark Node B As Visited And Include

The Edge (A,B)
Tree

i
4

d
f

Queue
Weight

0 6 4 5 3 4 2

1 2

Node

A B C D E F G H I

Predecessor

0 A D G C A F

D A
James Tam

Kruskals Algorithm: Description

Start out with a forest of unconnected trees (extract the
nodes from the graph)

Node

Node

Node

Becomes

Node

James Tam

Kruskals Algorithm: Description

Make the connections between the separate trees (add edges)
which have the lowest cost.
Connecting two separate nodes (two separate trees) merges
them into one tree

James Tam

Kruskals Algorithm
public Set kruskal (Graph g)
{
Set combinedSet = new Set ();
int edgesAccepted = 0;
PriorityQueue edgesLeft = g.sortEdges ();
Edge usedEdge;
Node sourceNode;
Node destinationNode;

James Tam

Kruskals Algorithm (2)

while (edgesAccepted < (g.getNumberNodes() - 1))
{
usedEdge = edgesLeft.dequeueMin ();
sourceNode = usedEdge.getSourceNode ();
destinationNode = usedEdge.getDestinationNode ();
if ((sourceNode != destinationNode) &&
(notCycle(sourceNode,destinationNode,combinedSet) == true)
{
edgesAccepted++;
combinedSet.union (sourceNode, destinationNode);
}
}
return combinedSet;
}
James Tam

Example Trace Of Kruskals Algorithm:

Original Graph
2

4
2

C
5

D
1

E
6

10

James Tam

Example Trace Of Kruskals Algorithm:

Sorted Edges In Priority Queue
Priority queue
2

C
5

D
1

E
6

10

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2
C - D: Weight = 2
D - E: Weight = 2
B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5
E G: Weight = 6
D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Forest Of Nodes
Priority queue
A

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2
C - D: Weight = 2
D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5
E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

First Edge Added (A-D)
Priority queue
A

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2
D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5
E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Second Edge Added (F-G)
Priority queue
A

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2
D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Third Edge Added (A-B)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2
D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Fourth Edge Added (C-D)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2
D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Add Fifth Edge (D-E)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2

D - E: Weight = 2

B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Dont Use Edge (B-D)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

C - D: Weight = 2
2

D - E: Weight = 2

Skip B - D: Weight = 3
A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Dont Use Edge (A-C)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

1
2

C - D: Weight = 2
2

D - E: Weight = 2

Skip B - D: Weight = 3
Skip A - C: Weight = 4
D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

Sixth Edge Added (D-G)
Priority queue
2

A - D: Weight = 1

F - G: Weight = 1
A - B: Weight = 2

1
2

C - D: Weight = 2
2

D - E: Weight = 2

Skip B - D: Weight = 3
Skip A - C: Weight = 4
D - G: Weight = 4

C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm: Stop, Edges

Accepted = No Nodes - 1
Priority queue
2

A
4

1
2

#1

A - D: Weight = 1

#2

F - G: Weight = 1

#3

A - B: Weight = 2

#4

C - D: Weight = 2

#5

D - E: Weight = 2

Skip B - D: Weight = 3
Skip A - C: Weight = 4

#6

D - G: Weight = 4
C F: Weight = 5

E G: Weight = 6

D F: Weight = 8
B E: Weight = 10
James Tam

Example Trace Of Kruskals Algorithm:

The Final Tree
2

G
James Tam

Topological Sorting Of Graphs: Real Life Example!

Used to order information that
could be represented as nodes in a
directed acyclic graph.
If theres a path from a node n1 to
another node n2, then n2 appears
after n1 in the ordering.

First year
Fall

Winter

CPSC 231 or
235

CPSC 233

Math 221

CPSC 265

Math 249 or
251

Math 271

Phil 279

Math 253 or
Stat 211
Second year

CPSC 325

CPSC 313

CPSC 331

SENG 311 or
CPSC 333

James Tam

Topological Sorting Of Graphs: Real Life Example

Used to order information that could be represented in the form
of the nodes in a directed acyclic graph:
Many seor
courses
CPSC 231

CPSC 265

CPSC 325

CPSC 233

CPSC 331

CPSC 235

SENG 311
Many seor
courses

James Tam

Note: Topological Sorting Cannot Be Done With

Cyclical Graphs
Sorry no vacancies today.1
Computer Science 231 H(3-2T-1)
Introduction to Computer Science I
Problem solving and programming in a
structured language. Data representation,
program control, basic file handling, the use of
simple data structures and their
implementation. Pointers. Recursion.
Prerequisite: Admission into the Department
of Computer Science

CPSC 231

Admission requirements into the

Department of Computer Science
A grade of C- or higher in CPSC 231

Admission into the

Department1

1 This example is purely fictional that was created to illustrate the principles of graph theory and should not be taken as an official
description of perquisite requirements for the Department of Computer Science

James Tam

Algorithm For A Topological Sort

public List topogicalSort (Graph g)
{
int i;
int noNodes = g.getNumberNodes ();
List orderedNodes = new List ();
Node temp;
for (i = 0; i < noNodes; i++)
{
temp = g.getNextTopParent ();
orderedNodes.add (temp);
g.deleteNodeEdges (temp);
}
}
James Tam

Example Of A Topological Sort

Graph

List

James Tam

Remove C And Edges From Graph And

Add To List

Graph

List
[0]
C

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove A

Graph

B
2

List
[0]

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove B

Graph

B
2

List
[0]

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove D

Graph

List
[0]

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove E

Graph

List
[0]

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove G

Graph

List
[0]

[1]

[2]

[3]

[4]

[5]

[6]
James Tam

Update The Numbers And Remove F

Graph

F
List
[0]

[1]

[2]

[3]

[4]

[5]

[6]

James Tam

You Should Now Know

What is a minimum spanning tree
Two algorithms (Prims and Kruskals) for creating minimum
spanning trees
What is a topological sort and the algorithm for this sort.

James Tam

Sources Of Lecture Material

Data Abstraction and Problem Solving With Java: Walls and
Mirrors updated edition by Frank M. Carrano and Janet J.
Prichard
Data Structures and Problem Solving Using C++ (2nd edition)
by Mark Allan Weiss
Data Structures and Problem Solving Using Java (2nd edition)
by Mark Allan Weiss
Lecture notes by Matthew A. Becker
http://www.andrew.cmu.edu/user/mbecker/
CPSC 331 course notes by Marina L. Gavrilova
http://pages.cpsc.ucalgary.ca/~marina/331/
James Tam