← → to navigate  |  T for contents

Table of Contents

Press T or Escape to close

1
DATA4100  ·  Week 4

Using IT Systems to Assist
Data Visualisation

Database modelling, Power BI data integration, and immersive analytics environments.

Analyse how IT assists data management and visualisation
Create database models and apply database queries
Visualise merged data in Power BI
Explore immersive environments and analytics
2

Where We Are — 13-Week Roadmap

1 Intro 2 Stats 3 Analytics 4 IT & DB You are here 5 Storytelling 6 Effective 7 Revision 8 Assessment 9 Adv. Vis 10 Transform 11 Python 12 Modelling Today's agenda covers three major areas: IT & Data Management Database Modelling + Power BI Immersive Analytics
3
Section 1 of 3

IT & Data
Visualisation

What is information technology, and how does it underpin modern data management and visualisation pipelines?

4

1.1 What is Information Technology?

Definition
Information Technology (IT) encompasses the physical devices, networks, and software processes used to store, process, secure, and transmit digital data.

Core IT domains

  • Internet & IoT — connected device ecosystems
  • Cloud storage & computing — remote, scalable infrastructure
  • Structured (SQL) & NoSQL databases — data organisation
  • AI & Machine Learning — automated insight generation
  • Cybersecurity — firewalls, VPNs, encryption
  • Data analytics tools — Power BI, Tableau, Python
  • IT Governance — policies for ethical data use
IT Systems Cloud & Storage SQL / NoSQL Cyber- security AI & ML Analytics Tools IT Governance
Figure 1.1: Core domains of Information Technology
5

1.2 How IT Assists Data Visualisation

IT DomainRole in Data Visualisation
Cloud storage & computingData preparation pipelines; remote access to large datasets for visual analysis
SQL / NoSQL databasesExtract precisely the rows and columns you need — query before you visualise
AI & Machine LearningAutomated chart recommendations, natural-language queries, pattern detection
CybersecurityKeeps dashboards and company data secure from unauthorised access
Analytics tools (Power BI, Tableau)Build interactive, shareable dashboards and immersive visualisations
IT GovernanceEnsures ethical data collection, privacy compliance, and audit trails
Key Insight
IT is not just infrastructure — it shapes what you can visualise, how fast you can do it, and who can access the result.
6

Activity 1 — How Does IT Assist Visualisation?

Group Activity — 8 minutes
  1. Form groups of 3–4.
  2. List three ways IT assists data management and visualisation in your chosen industry (health, retail, finance, education, etc.).
  3. For each way, name: the IT type, how it helps, and one risk if it failed.
  4. Report back to the class.
Example — Retail
  • Cloud database stores real-time inventory across 200 stores → visualise stock levels instantly
  • SQL query filters to low-stock SKUs → focused reorder dashboard
  • AI flags demand anomalies → proactive restocking chart

Prompt questions

  • What data is being collected?
  • Where is it stored?
  • Who needs to see the visualisation?
  • What security concerns exist?
  • What happens if the IT system fails?
7

Knowledge Check — Section 1

Q1. A data analyst wants to extract only sales records from Queensland between January and March. Which IT domain is most directly responsible for enabling this?
SQL allows analysts to filter rows precisely using WHERE clauses — this is the primary mechanism for subsetting data before visualisation.
Q2. Power BI's Copilot feature lets users ask data questions in plain English. Which IT domain does this fall under?
Natural-language query interfaces are powered by large language models — a core application of AI/ML within analytics tools.
8
Section 2 of 3

Database Modelling
& Power BI

How data is structured in relational tables, how tables are connected through keys, and how to merge and visualise multi-table data in Power BI.

9

2.1 What is Database Modelling?

Definition
Database modelling is the process of creating a visual and structural representation of how data is organised in a system — what data exists, how it is grouped, and how different groups relate to each other.

Why it matters

  • Prevents duplicate or inconsistent data
  • Makes queries faster and more reliable
  • Enables accurate cross-table analysis and reporting
  • Informs dashboard design — if the model is wrong, the visual is wrong

Four key components

  • Tables — containers for related data
  • Columns (attributes) — properties of each record
  • Relationships — how tables are connected
  • Keys — unique identifiers linking tables together
customers 🔑 CustomerID CustomerName Email PhoneNumber products 🔑 ProductID ProductName Price orders 🔑 InvID FK CustomerID FK ProductID QTY 1 * 1 *
Figure 2.1: A simple entity-relationship model
10

2.2 Key Database Design Terms

Entity
A person, place, thing, or event about which data is collected. Represents a single, unique item.
Example
A customer, a product, a staff member
Attribute
A characteristic that describes an entity — stored as a column in a table.
Example
CustomerID, FirstName, DateOfBirth, Email
Relationship
A logical link between two entities, typically through a shared key. Documented in an Entity-Relationship (ER) diagram.
Example
A customer places many orders (1-to-many)
Primary Key (PK)
A unique identifier for each row in a table. No two rows can have the same PK. Example: CustomerID = CUST001
Foreign Key (FK)
A column in one table that references the PK of another table — this creates the relationship. Example: orders.CustomerID → customers.CustomerID
11

2.3 Relational Table Structure

In a relational database, data is organised into tables (also called relations). Every table follows the same basic structure:

FirstNameLastNamePayrollIDYTD Tax
JuliaWangP123414$5,693
BobJonesP231118$13,004
HarpreetSinghP284923$8,934
RosemaryChuP285026$10,567
Database Query
A structured request to retrieve specific rows or columns from one or more tables. SQL is the most common language. Example: "Give me all orders by customers in Queensland in Q1 2024."
Power BI Connection
Tables like Excel/CSV files are connected via a shared column (e.g., StudentID). Power BI can run queries and automatically detect these relationships.
12

2.4 Business Scenario — Retail Sales Analysis

Scenario
You are a data analyst for a retail company. You have been asked to establish relationships among data tables and generate reports on sales performance, customer behaviour, product popularity, and staff contributions.

You are given four CSV tables:

customers 🔑 CustomerID CustomerName Email PhoneNumber products 🔑 ProductID ProductName Price staff 🔑 StaffID StaffName Email HireDate orders 🔑 InvID FK CustomerID FK ProductID FK StaffID QTY Date
Figure 2.2: Four CSV tables linked via Foreign Keys into the Orders fact table
13

2.5 Power BI — Loading the Four Tables

Goal
Import all four CSV files (customers, products, staff, orders) into Power BI and set column headers correctly.
  1. 1
    Open Power BI Desktop. On the home screen, click Get data from another source → select Text/CSV.
  2. 2
    Navigate to your customers.csv file, click Load. Power BI shows a preview — check the delimiter is set to Comma.
  3. 3
    Repeat for products.csv, staff.csv, and orders.csv. After all four loads, confirm four table names appear in the Data pane (right side).
  4. 4
    Click Transform Data (top ribbon → Queries group) to open the Power Query Editor.
  5. 5
    For each table: click Use First Row as Headers. This promotes your CSV header row into proper column names. Repeat for all four tables.
  6. 6
    Click Close & Apply to load the transformed tables back into Power BI.
14

2.6 Power BI — Viewing & Verifying Relationships

Power BI often auto-detects relationships when column names match across tables. You should always verify this.

  1. 1
    Click the Model view icon (left sidebar — looks like a diagram).
  2. 2
    You should see four table cards connected by lines. Each line = one relationship.
  3. 3
    Click each relationship line to inspect it in the Properties pane: confirm the correct columns are linked (e.g., orders.CustomerID → customers.CustomerID).
  4. 4
    If a relationship is missing: drag the column name from the lookup table onto the matching column in the orders table.
  5. 5
    Check the Cardinality — it should be Many to one (*:1) — many orders map to one customer/product/staff.
Three relationships to verify
orders.CustomerID → customers.CustomerID
orders.ProductID → products.ProductID
orders.StaffID → staff.StaffID
customers 🔑 CustomerID products 🔑 ProductID staff 🔑 StaffID orders FK CustomerID FK ProductID FK StaffID 1 * 1 * * 1
Figure 2.3: Star schema — Orders as the fact table
15

2.7 Power BI — Merging Queries & Adding Calculated Columns

Step A — Bring Price into Orders (Merge)

  1. 1
    Open Transform Data → select the orders query in the left panel.
  2. 2
    Click Merge Queries (Home ribbon, Combine group).
  3. 3
    Select products as the second table. Click ProductID in both tables to link them. Join kind: Left Outer. Click OK.
  4. 4
    A new products column appears. Click the expand icon → tick only Price. Untick "Use original column name as prefix". Click OK.

Step B — Calculate Total Sales

  1. 1
    Hold Ctrl and select both the QTY and Price columns.
  2. 2
    Right-click → Transform → Product. Power Query multiplies QTY × Price row by row.
  3. 3
    Rename the new column to Total Sales by double-clicking the column header.
  4. 4
    Click Close & Apply to load back into Power BI.
Result
The orders table now contains a Total Sales column ready to use in visualisations — no DAX measure needed at this stage.
16

2.8 Power BI — Creating Visuals from Merged Data

With the data model complete, you can now build visuals that span all four tables. Here are three visuals to build:

Visual 1 — Summary Table
Fields: staff.StaffName, Sum of QTY, Sum of Total Sales

Shows each staff member's sales volume and revenue — useful for performance review.
Visual 2 — Bar Chart
X-axis: Sum of Total Sales
Y-axis: customers.CustomerName

Ranks customers by revenue — identifies your top buyers.
Visual 3 — Column Chart
X-axis: products.ProductName
Y-axis: Sum of Total Sales

Shows revenue by product — highlights best-sellers and slow movers.
Why this matters
None of these visuals would be possible from a single CSV. The data model — the relationships and merge — is what makes cross-table analysis possible. Good modelling → richer visualisation.
17

Knowledge Check — Section 2

Q1. In the retail scenario, the orders table contains a CustomerID column. In database terms, what role does this column play?
A Foreign Key in the orders table references the Primary Key in the customers table, establishing the one-to-many relationship: one customer can have many orders.
Q2. After merging the products table into orders and creating a "Total Sales" column, you now have 8 columns in orders. What is the correct formula Power Query applied?
The "Product" transform in Power Query multiplies two columns row-by-row. QTY × Price gives the revenue for each individual order line.
18
Section 3 of 3

Immersive Environments
& Analytics

From standard 2D dashboards to virtual and mixed reality — how IT enables analysts to step inside their data.

19

3.1 What are Immersive Environments?

Virtual Reality (VR)
A realistic and immersive simulation of a three-dimensional environment, created using interactive software and hardware, and experienced or controlled by movement of the body. The user is fully enclosed in the virtual world.
Augmented Virtuality (AV)
Computer-generated imagery that can be manipulated by the user — e.g., designing a kitchen layout by dragging virtual components on a touchscreen.
Mixed Reality (MR)
Augmenting the real environment with digital elements. The real world remains visible but is overlaid with virtual objects. Example: Pokémon GO, Microsoft HoloLens.
Real Physical world MR Mixed Reality AV Augmented Virtuality VR Virtual Reality ← More real world More virtual → Reality–Virtuality Continuum (Milgram & Kishino, 1994)
Figure 3.1: The Reality–Virtuality Continuum
All three are Immersive Environments (IE)
They share the goal of placing the user inside an experience, rather than observing it from outside on a screen.
20

3.2 Business Scenario — Beyond Static Dashboards

Scenario
You are a data visualisation analyst. You have been reporting to and training staff using static and interactive 3-D visuals — but you have found this is often inadequate for complex spatial or operational data. You want to go one step further: create an illusionary environment that your colleagues can step into and interact with directly.

Which industries need this?

  • Health — surgical training, patient anatomy
  • Resources — mine site hazard simulation
  • Retail — virtual store layout testing
  • Communications — network infrastructure walkthroughs
  • Education — field trips, laboratory simulations
  • Technology — hardware design reviews

Discussion prompt

For your industry, answer:

  • What would colleagues need to see in the IE?
  • What data drives the environment?
  • What decisions would they make inside it?
  • What hardware/software would you choose?
21

3.3 Case Study 1 — Student Learning via Imaginarium

Business Problem
Coomera Anglican College was struggling to engage students through traditional teaching methods.
Solution — Immersaview Imaginarium
A climate-controlled, 360° room was built featuring six laser projectors and cinema-quality surround sound — delivering a seamless sensory experience without wearable technology.
View the Imaginarium
https://youtu.be/94ga2HkV0KU

How realistic are these environments? What data do you think powers them?

6 Laser Projectors 360° seamless projection
Figure 3.2: Imaginarium schematic — full-room immersive display
22

3.4 Case Study 2 — Helicopter Rescue Training (Varjo)

Business Problem
Rescuing people from road-inaccessible locations requires specific physical skills — but real-world training is expensive, dangerous, and logistically difficult.
Solution — Varjo VR
Varjo Technologies built a VR simulation that lets rescuers practise attaching an injured person to rescue equipment and pulling them up to a helicopter — repeatedly, safely, at a fraction of the cost.
Watch the Simulation
youtube.com/shorts/qz0Xcb7UWhk

Why is VR training effective here?

  • Safety — no real-world risk during skill development
  • Repetition — trainees can repeat scenarios indefinitely
  • Data capture — every movement is logged; performance is visualised
  • Cost — one VR setup replaces expensive physical drills
  • Standardisation — every trainee faces identical scenarios
Data Visualisation Link
VR training platforms generate motion, timing, and decision data — which is then visualised to identify skill gaps and measure improvement over time.
23

3.5 Considerations When Creating an Immersive Environment

1. Attention to Detail
  • List every required component (hardware, software, content)
  • Create sensory richness — visual fidelity, audio, haptics where possible
  • Inconsistencies break immersion
2. Audience Transformation
  • How will the environment change the user's understanding or behaviour?
  • Define the role each participant takes in the simulation
  • Design for the decision they need to practise
3. Technical Design
  • Assess the physical space available
  • Allow the user to make meaningful decisions inside the IE
  • Choose hardware/software that is fit for purpose — not just the most impressive
Design Challenge
Choose one industry from the earlier list. Sketch a brief IE concept: What does the user see? What data powers it? What decision do they practise? What hardware would you use?
24

3.6 Immersive Analytics

Immersive Analytics
The application of VR or mixed reality to data exploration — enabling analysts to literally walk through their data, rather than viewing it on a flat screen.
Situated Analytics
A subset where virtual data is spatially organised in relation to real-world objects, locations, and people.
  • Virtual price tags overlaid on physical shelves
  • IoT alerts appearing above machines on a factory floor
  • Patient vitals displayed above a hospital bed

Leading platforms

  • Virtualitics — AI-powered 3D data exploration in VR
  • Immersion Analytics — MagicLeap and HoloLens integration
  • Handbuilt / Zigfu — gesture-controlled industrial data
Traditional 2D Dashboard Immersive Analytics Immersive Environment Benefits of Immersive Analytics → Spatial pattern recognition in 3D → Natural interaction (gesture, gaze) → Collaborative multi-user environments → Situated, contextual data display → Higher engagement and retention
Figure 3.3: From flat dashboard to immersive data space
25

Activity 2 — VR in Healthcare

Group Activity — 10 minutes
Watch: https://www.youtube.com/watch?v=uCQU6g1J4HM
(Applications of VR in Healthcare — 5:15)

Q1. What applications are shown?

List the specific healthcare contexts where VR is being used. What problems does each application solve?

Q2. What data is required?

For each application, what underlying data drives the environment? Think about: patient records, imaging data, motion capture, physiological signals.

Q3. How useful are these?

What are the concrete benefits over traditional training/treatment? What limitations or risks might exist? Is the data visualisation aspect clear?

Connect to Course Themes
Notice how immersive analytics is still fundamentally about data — the VR is just a new display medium for that data. The same principles of clarity, accuracy, and audience-awareness still apply.
26

Activity 3 — Immersive Analytics in Manufacturing

Group Activity — 8 minutes
Watch: https://www.youtube.com/watch?v=VwbG20pvPzY
(Immersion Analytics in Manufacturing — 2:51)
Q1 — What data?
What sort of data is the user interacting with in this case? Consider: spatial data, time-series sensor readings, equipment status, production metrics.
Q2 — Advantage over standard?
Why is immersive analytics advantageous here compared to a standard 2D Power BI dashboard? What specifically is gained by the 3D spatial context?
Q3 — Other applications?
Can you think of another manufacturing, logistics, or infrastructure context where this technology would add clear value? What data would it need?
27

Knowledge Check — Section 3

Q1. A warehouse manager wears AR glasses that show real-time stock levels floating above each shelf location. Which concept best describes this?
Situated Analytics places virtual data in relation to real-world objects and locations. The manager can see the physical shelf AND the data associated with it simultaneously — this is anchoring data to place, not full immersion.
Q2. In the helicopter rescue VR case study, what is the key data visualisation benefit beyond the training simulation itself?
VR training environments are also data collection environments. The performance data they generate — timing, movement accuracy, decision sequences — is then visualised to guide instructor feedback and trainee improvement.
28

Lesson Summary — Week 4

Section 1 — IT & Vis

  • IT = hardware + software + networks for managing digital data
  • Six IT domains directly support visualisation: cloud, databases, AI/ML, cybersecurity, analytics tools, governance
  • IT shapes what you can visualise and who can access it

Section 2 — Databases & Power BI

  • Relational databases organise data into linked tables via Primary and Foreign Keys
  • Power BI can load, transform, relate, and merge tables
  • Calculated columns (e.g., Total Sales = QTY × Price) enable richer visualisation
  • Good data modelling → accurate, cross-table dashboards

Section 3 — Immersive

  • VR (fully virtual), MR (overlaid on real), AV (manipulable CGI)
  • Immersive Analytics brings data into 3D, navigable space
  • Situated Analytics anchors data to real-world objects
  • Applications: education, rescue training, healthcare, manufacturing
Looking Ahead — Week 5
Communication of visualisations and storytelling — how to present data insights persuasively to a business audience.