INFS7450 SOCIAL MEDIA ANALYTICS

Week 2

Liang QU
School of EECS
The University of Queensland
About me: Liang QU

• Office room: 78-622

• Email: liang.qu@uq.edu.au
• Face-to-face: appointment by email
• Research interests: graph embedding, recommender systems, LLM
Today’s Contents

• About INFS7450
• Graph Essentials
• Coding Demo
• Q&A
Section 1: Recall Key Points on INFS7450
Section 1: Course Schedule
Section 1: Course Website - Blackboard

• Lectures, Tutorials and Their Recordings

– Slides posted at the night before the lecture and tutorial sessions
respectively.
– Their recordings will be available on Blackboard system.
• Assignments
– Both online quizzes and projects will be released and submitted via
Blackboard system.
• Communication
– All announcements and reminders about the deadline and due date will be
posted through Blackboard.
– Ed Discussion Board will be used for discussions, queries and questions.
• Emails to the tutors for private questions; all other questions should be posted to ED.
• Students are encouraged to help each other on the discussion board.
• If you cannot access this course on Ed Discussion, please contact
Skye(skye.sun@uq.edu.au )
Section 1: About Generative AI

Examples of misuse:
• Reports: Submitting a report generated by AI without indicating that AI was used.
• Code: Incorporating AI-generated code in your projects without citing the AI as a source.
• Quizzes: Using AI to answer quiz questions

Failure to properly cite AI contributions is a form of academic misconduct

Always attribute AI assistance in your work

Course Profiles: https://course-profiles.uq.edu.au/student_section_loader/section_1/132633

Section 2: Graph Essentials
Section 2: In this section, we’d like to answer:
• Types of graphs
– Directed graph VS undirected graph
– Weighted graph VS unweighted graph
– Regular graph VS complete graph
– Simple graph VS multigraph
• How to traverse a graph?
– DFS
– BFS
• What is a strongly connected component in a
directed graph?
• How to find the shortest path in a weighted graph?
Section 2: Types of Graphs
• Directed graph VS undirected graph
– Handshake Theorem: Let G=(V,E) be an undirected graph with m edges. Then
2𝑚 = $ deg(𝑣)
!∈#
Proof: Each edge contributes twice to the total degree count of all vertices. Thus, both sides
of the equation equal to twice the number of edges.
– Theorem: An undirected graph has an even number of nodes of odd degree.
Proof: Let V1 be the nodes of even degree and V2 be the nodes of odd degree in an
undirected graph G=(V,E) with m edges. Then
2𝑚 = ∑!∈# deg(𝑣) = ∑!∈#! deg 𝑣 + ∑!∈#" deg(𝑣)

Directed Undirected
Section 2: Types of Graphs
• Weighted graph VS unweighted graph
– Degree has generally been extended to the sum of weights when analyzing
weighted graph.
Section 2: Types of Graphs
• Regular graph VS complete graph
– A regular graph is a graph where each node has the same number of
neighbors; i.e., every node has the same degree.
– A complete graph is a graph in which each pair of graph nodes is connected
by an edge.
– A complete graph is always a regular graph.

3-regular graph Complete graph

Section 2: Types of Graphs
• Simple graph VS multigraph
– Simple graphs are graphs where only a single edge is allowed to exist
between any pair of nodes (contain no self loops)
– Multigraphs are graphs where you can have multiple edges between two
nodes (loops are allowed)
 Section 2: How to traverse a graph?

• Graph traversal refers to the process of visiting each node in

a graph.
• Typical solution:
– Depth First Search (DFS)
– Breath First Search (BFS)

1 6

0 2
4
7
3
5
 Section 2: DFS

• The Depth First Search (DFS) is the most fundamental search

algorithm used to explore nodes and edges of a graph.
• As the name suggests, a DFS plunges depth first into a graph
without regard for which edge it takes next until it cannot go
any further at which point it backtracks and continues.

1 6

0 2
4
7
3
5
Section 2: DFS

• Start DFS at node 0

• The node 0 is in the stack.
1 6

0 2
4
7
3
5

0 The order of visited nodes: {}

Section 2: DFS

• The node 0 is out of stack

• Put the neighbor nodes of node 0 to the stack

1 6

0 2
4
7
3
5

0 The order of visited nodes: {0}

Section 2: DFS

• The node 0 is out of stack

• Put the neighbor nodes of node 0 to the stack

1 6

0 2
4
7
3
5

0 The order of visited nodes: {0}

Section 2: DFS

• Stack follows the LIFO (Last in First out) principle

• Node 1 is going out of the stack

1 6

0 2
4
7
3
5

0 The order of visited nodes: {0}

Section 2: DFS

• Node 1 is out of the stack

• Put the neighbor node of node 1 into the stack

1 6

0 2
4
7
3
5
2

0 The order of visited nodes: {0,1}

Section 2: DFS

• Stack follows the LIFO (Last in First out) principle

• Node 2 is going out of the stack

1 6

0 2
4
7
3
5
2

0 The order of visited nodes: {0,1}

Section 2: DFS

• Node 2 is out of the stack

• Put the neighbor node of node 2 into the stack

1 6

0 2
4
7
4 3
5
2

0 The order of visited nodes: {0,1,2}

Section 2: DFS

• Stack follows the LIFO (Last in First out) principle

• Node 4 is going out of the stack

1 6

0 2
4
7
4 3
5
2

0 The order of visited nodes: {0,1,2}

Section 2: DFS

• Node 4 is out of the stack

• Put the neighbor node of node 4 into the stack

1 6

0 2
6 4
7
4 3
5
2

0 The order of visited nodes: {0,1,2,4}

Section 2: DFS

• Node 4 is out of the stack

• Put the neighbor node of node 4 into the stack

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

0 The order of visited nodes: {0,1,2,4}

Section 2: DFS

• Node 4 is out of the stack

• Put the neighbor node of node 4 into the stack

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

0 The order of visited nodes: {0,1,2,4}

Section 2: DFS

• Stack follows the LIFO (Last in First out) principle

• Node 5 is going out of the stack

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

0 The order of visited nodes: {0,1,2,4}

Section 2: DFS
• Node 5 is out of the stack
• Put the neighbor node of node 5 into the stack
• Node 7 is already in the stack

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

0 The order of visited nodes: {0,1,2,4,5}

Section 2: DFS

• Stack follows the LIFO (Last in First out) principle

• Node 7 is going out of the stack

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

0 The order of visited nodes: {0,1,2,4,5}

Section 2: DFS
• Node 7 is out of the stack
• Backtrack the stack
• Make sure you don’t re-visit visited nodes!

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

0 The order of visited nodes: {0,1,2,4,5,7}

Section 2: DFS
• Node 6 is going out of the stack
• Backtrack when a dead end is reached.

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

0 The order of visited nodes: {0,1,2,4,5,7}

Section 2: DFS
• Node 6 is out of the stack
• Backtrack

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

0 The order of visited nodes: {0,1,2,4,5,7,6}

Section 2: DFS
• Node 3 is going out of the stack

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

0 The order of visited nodes: {0,1,2,4,5,7,6}

Section 2: DFS
• Node 3 is out of the stack
• The DFS will terminate when the stack is empty
7
5
1 6
7
0 2
6 4
7
4 3
5
2

0 The order of visited nodes: {0,1,2,4,5,7,6,3}

Section 2: Strongly Connected Component

• What is a strongly connected component?

– A strongly connected component is the portion of a directed graph in which there is a
path from each node to another node.
– It is applicable only on a directed graph.
– Is {a,b,c} a strongly connected component?
– Is {b,c,d} a strongly connected component?

– How to find strongly connected components? You can refer to:

https://stackoverflow.com/questions/33590974/how-to-find-strongly-connected-
components-in-a-graph
Section 2: Breadth First Search (BFS)

• The Breadth First Search (BFS) is another fundamental

search algorithm used to explore nodes and edges of a
graph.

1 6

0 2
4
7
3
5
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5
Section 2: Using a Queue

• The BFS algorithm uses a queue data structure to track which

node to visit next. Upon reaching a new node the algorithm adds
it the queue to visit it later. The queue data structure works just
like a real-world queue such as a waiting line at a restaurant.
People can either join the waiting line (enqueue) or get seated
(dequeue)
Queue Back
Dequeue
Enqueue

Queue Front
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
3
5

0
Section 2: Breadth First Search (BFS)

• Add the neighbors of node 1 to the queue.

1 6

0 2
4
7
3
5

0
Section 2: Breadth First Search (BFS)

• Add the neighbors of node 1 to the queue.

1 6

0 2
4
7
3
5
3

0
Section 2: Breadth First Search (BFS)

• After we add all the neighbor nodes of node 0 to the queue, we say that we
have visited the node 0.
• The node 0 will dequeue.
• We will visit the next node in the queue, i..e, the node 1

1 6

0 2
4
7
3
5
3

0
Section 2: Breadth First Search (BFS)

• We add the neighbor node (node 2) of node 1 to the queue.

• Node 0 has been visited, so we don’t add it to the queue.

1 6

0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• After we add all the neighbor nodes of node 1 to the queue, we say that we have
visited the node 1.
• The node 1 will dequeue.
• We will visit the next node in the queue, i..e, the node 3
• Here is the difference between DFS and BFS

1 6

0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)

• A BFS starts at some arbitrary node of a graph and explore

the neighbor nodes first, before moving to the next level
neighbors
5

7
1 6
6

4 0 2
4
7
2 3
5
3

0
Section 2: Breadth First Search (BFS)-Application

• The BFS algorithm is particularly useful for one thing:

finding the shortest path on weight/unweighted graphs.
• What is the shortest path length between node 0 and node
4?
3 1 7
0 2 4
4
2 1
3
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
4
2 1
3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
4
2 1
3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
4
2 1
3 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
4
2 1
3 3

1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
2 4
2 1
3 3

1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
2 4
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
2 4
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
0 2 4
Path: 0 à1à2, Length: 3+7=10
2 4
Path: 0 à3à2, Length:2+1=3
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

3 1 7
4
0 2 4
Path: 0 à1à2, Length: 3+7=10
2 4
Path: 0 à3à2, Length:2+1=3
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

Path: 0 à1à2à4, Length: 3+7+4=14

3 1 7
4 Path: 0 à3à2à4, Length:2+1+4=7
0 2 4
Path: 0 à1à2, Length: 3+7=10
2 4
Path: 0 à3à2, Length:2+1=3
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
Section 2: Breadth First Search (BFS)-Application

• What is the shortest path length between node 0 and node

4?

Path: 0 à1à2à4, Length: 3+7+4=14

3 1 7
4 Path: 0 à3à2à4, Length:2+1+4=7
0 2 4
Path: 0 à1à2, Length: 3+7=10
2 4
Path: 0 à3à2, Length:2+1=3
2 1
3 3
Path: 0 -->3, Length: 2
1 Path: 0 -->1, Length: 3

0
A: BFS

B: DFS

C: Neither BFS and DFS will ever

encounter the goal node in this graph

D: BFS and DFS encounter same number

of nodes before encounter the goal node
Section 3: Coding Demo
Section3: Coding demo
• Configure your development environment:

• Install Anaconda on Windows:

• https://docs.anaconda.com/anaconda/install/windows/

• Install Anaconda on MacOS:

• https://docs.anaconda.com/anaconda/install/mac-os/

• Install Jupyter Notebook (Optional):

• https://docs.jupyter.org/en/latest/install/notebook-classic.html
 See you next weekJ

Don’t forget the Online quiz 1!