Foundations of Data Visualisation - Computerphile

TL;DR

Validation in visualization is a complex and subjective process, with no universally accepted framework. It involves mapping attributes of data items to visual representations using marks and channels.

Transcript

today we will talk a bit more about the kind of foundations of graduation in terms of kind of a more simplistic or theoretical way of thinking of validation and how to use that to design your next validation so this is not the teacher how to use spreadsheets to create a bar chart but to say what makes a good pie chart the point I want to make here ... Read More

Key Insights

• 👤 Validation in visualization is subjective and depends on individual interpretation, data, and user cases.
• 😵 Visualization is a cross-disciplinary field, drawing from various disciplines like computer science, psychology, and neuroscience.
• 💋 Marks represent data items, while channels represent attributes of the data items.
• 🗯️ Choosing the right channels for visualization requires experimentation and comparing their effectiveness.
• 💠 Position (X, Y) is an effective channel, while 3D shapes are generally less recommended.
• 🎚️ Different channels have varying levels of accuracy in representing data attributes.
• ❓ People tend to overestimate certain channels and underestimate others.

Questions & Answers

Q: How is validation in visualization different from other fields?

Validation in visualization is a cross-disciplinary field, combining computer science, computer graphics, human-computer interaction, psychology, and neuroscience. It involves creating visual representations using marks and channels.

Q: What are marks and channels in visualization?

Marks are data items, such as products in a spreadsheet, represented visually. Channels are the visual representations of attributes of the data items, such as color, shape, size, or position.

Q: How do you choose the appropriate channels for visualization?

Choosing the right channels for visualization requires experimentation and comparing their effectiveness. Position (X, Y coordinates) and color are often effective, while 3D shapes are generally less recommended.

Q: How accurate are different channels in representing data attributes?

Different channels have varying levels of accuracy in representing data attributes. Position (X, Y) is highly effective, while 3D shapes are less accurate. Experimentation has shown that people tend to overestimate certain channels and underestimate others.

Summary & Key Takeaways

• Validation in visualization is like an art with personal interpretation and depends on data and user cases.

• Visualization is a cross-disciplinary field, drawing from computer science, computer graphics, human-computer interaction, psychology, and neuroscience.

• Visualization involves using marks (data items) and channels (visual representation of attributes) to create visual representations.