DCIT422: Information
Visualization
Session 1 – Introduction
By
Solomon Mensah (PhD)
College of Education
School of Continuing and Distance Education
2014/2015 – 2016/2017
Course Outline
• What is Information Visualization?
• Historical Background
• Principles of Effective Visualization
• Types of Visualizations
• Tools & Libraries for Visualization
• Importance & Value in a Data-Driven World
• Applications Across Domains
• Key Concepts to Explore
• Challenges & Future in Information Visualization
• Assignment I
Slide 2
What is Information Visualization?
Definition:
Information visualization is the graphical representation of data to
facilitate understanding, exploration, and communication of insights. It
combines principles from Data science, Design & Art, Cognitive
psychology & Human-computer interaction (HCI).
Key Characteristics:
• Transforms complex datasets into intuitive visuals (e.g., charts,
maps, graphs).
• Emphasizes clarity, accuracy, and interactivity.
• Supports human cognition by leveraging visual perception to
identify patterns, trends, and outliers.
[Link] Slide 3
visual%20display%20of%20quantitative%[Link]
Example: A line chart showing global temperature changes over
time or a heat map of website user activity
[Link] Slide 4
temperature-report-for-2023/
Example: A line chart showing global temperature changes over
time or a heat map of website user activity
This graph titled “Global Warming 1850 to 2023” shows the global temperature
anomaly (change in global temperature relative to the 1850–1900 average) based on
multiple independent datasets:
• HadCRUT5 (blue)
• GISTEMP (orange)
• NOAA (yellow)
• ECMWF (purple)
• Berkeley Earth (black line with grey 95% confidence interval)
Interpretation:
• All datasets consistently show a long-term warming trend, especially after the
1970s.
• The temperature anomaly has increased by over 1.4°C as of 2023.
• Despite some short-term fluctuations (e.g., dips in early 20th century), the overall
trajectory is clearly upward.
Slide 5
Historical Background
Key Milestones:
• 1801: William Playfair invents the bar chart and pie chart,
pioneering data visualization.
• 1854: John Snow’s cholera map identifies disease spread,
demonstrating visualization’s impact.
• 1970s: Edward Tufte’s work emphasizes data-ink ratio and
clarity in visualization design.
• 2000s: Rise of interactive and web-based visualizations with
tools like [Link].
• 2010s-Present: Big data and AI fuel advanced visualizations,
from real-time dashboards to 3D models.
[Link] Slide 6
nformation_visualization
Principles of Effective Visualization
Core Principles:
• Clarity: Avoid clutter, use clear labels, and prioritize data
over decoration.
• Accuracy: Represent data truthfully, avoiding distortion
(e.g., misleading axis scales).
• Efficiency: Maximize data-ink ratio (use ink for data, not
embellishment).
• Interactivity: Allow users to explore data (e.g., zoom, filter)
when appropriate.
• Audience Focus: Tailor visuals to the audience’s needs and
expertise.
[Link] Slide 7
20visual%20display%20of%20quantitative%[Link]
Types of Visualizations
Common Types:
• Charts: Bar, line, pie, scatter (e.g., stock price trends).
• Maps: Geographic (e.g., heatmaps of population density).
• Graphs: Network diagrams (e.g., social network connections).
• Dashboards: Combine multiple visuals for real-time insights.
• Infographics: Combine visuals and text for storytelling.
Choosing the Right Type:
• Bar charts for comparisons, line charts for trends, maps for
spatial data.
• Consider data type (categorical, numerical) and audience needs.
Slide 8
Tools & Libraries for Visualization
[Link]:
• JavaScript library for creating interactive, web-based
visualizations.
• Highly customizable, ideal for developers.
• Example: Interactive network graphs or animated charts.
Tableau:
• User-friendly platform for creating dashboards and
reports.
• Drag-and-drop interface, suitable for non-programmers.
• Example: Business intelligence dashboards.
Slide 9
Tools & Libraries for Visualization
Python Libraries:
• Matplotlib: Flexible library for static and interactive plots.
• Seaborn: Built on Matplotlib, simplifies statistical
visualizations.
• Plotly: Supports interactive and web-based visuals.
• Example: Heatmaps, scatter plots for data analysis.
Other Tools:
• Power BI, Google Data Studio, R’s ggplot2.
Slide 10
Importance & Value in a Data-Driven World
Importance:
• Simplifies Complexity: Makes large, complex datasets
understandable (e.g., billions of data points in climate
models).
• Reveals Patterns: Highlights trends, correlations, and
anomalies that might be missed in raw data.
• Supports Decision-Making: Enables stakeholders to make
informed choices in business, policy, and research.
• Enhances Communication: Conveys insights to diverse
audiences, from experts to the public.
Slide 11
Importance & Value in a Data-Driven World
Value:
• Increases engagement: Visuals are more memorable than
text or numbers.
• Reduces cognitive load: Humans process visuals 60,000 times
faster than text ([3M Corporation Study, 1986]).
• Drives action: Clear visuals persuade and inform (e.g., public
health dashboards during pandemics).
Slide 12
Applications Across Domains
Artificial Intelligence (AI):
• Visualizing model performance (e.g., loss curves, confusion
matrices).
• Explaining complex models (e.g., feature importance in neural
networks).
• Example: TensorBoard for visualizing machine learning metrics.
Journalism:
• Data-driven storytelling (e.g., election maps, climate change
charts).
• Example: The New York Times’ interactive visualizations of
COVID-19 spread.
Slide 13
Applications Across Domains
Civic Tech:
• Empowering citizens with transparent data (e.g., budget
visualizations, crime maps).
• Example: Open government dashboards showing public
spending.
Other Domains:
• Healthcare: Visualizing patient data or disease trends.
• Business: Sales dashboards, market trend analysis.
• Science: Visualizing experimental results (e.g., particle
physics data).
Slide 14
Key Concepts to Explore
Key concepts to explore in this Information Visualization:
• Visual encoding and perceptual design
• Data abstraction and transformation
• Interactivity and storytelling
• Visualizing high-dimensional, textual, and network
data
• Design evaluation and critique
Slide 15
Challenges in Information Visualization
• Data Quality: Incomplete or noisy data can lead to
misleading visuals.
• Design Choices: Poor choices (e.g., 3D effects, too many
colors) reduce clarity.
• Accessibility: Visuals must be usable by all, including
colorblind users.
• Scalability: Handling large datasets without overwhelming
the viewer.
• Ethical Considerations: Avoiding bias or misrepresentation
in data presentation.
Slide 16
Future of Information Visualization
Emerging Trends:
• Interactive Visualizations: Real-time exploration with web-
based tools.
• AI-Driven Visualizations: Automated insights and dynamic
visuals (e.g., Tableau’s AI features).
• Augmented Reality (AR)/VR: Immersive data exploration (e.g.,
3D data models).
• Storytelling: Combining visuals with narrative for impact.
Ethical Considerations:
• Ensuring transparency in data sources.
• Designing for inclusivity and accessibility.
Slide 17
Conclusion
• Information visualization transforms data into insights, driving
decisions and communication.
• Applications span AI, journalism, civic tech, and beyond.
• Tools like [Link], Tableau, and Python libraries empower
creators.
• Assignment 1 fosters critical analysis of real-world
visualizations.
Slide 18
Assignment 1: Visualization Critique and
Reflection
1. Objective: Analyze a real-world visualization and reflect on its
effectiveness.
2. Tasks:
• Find a Visualization: Choose a visualization from a reputable
source (e.g., The New York Times, Our World in Data, or a public
dashboard).
• Critique (1-2 pages):
§ Describe the visualization (type, data, purpose).
§ Evaluate based on principles (clarity, accuracy, efficiency,
audience focus).
§ Identify strengths and weaknesses.
§ Suggest improvements (e.g., better chart type, less clutter).
Slide 20
Assignment 1: Visualization Critique and
Reflection
3. Reflection (1 Page):
• What did you learn about effective visualization?
• How might you apply these principles in your work?
4. Submission:
• Written report.
Slide 21
Thank you
