THINK FIRST·CODE LATER

← Interactive Information Visualization
Chapter 1 · Week 1

Visualization in the Age of AI

Answered 0/30 Correct 0
Sign in to save progress across devices
Q1

According to Munzner's definition, what is the purpose of a computer-based visualization system?

Q2

When is a visualization least needed?

Q3

The four datasets of Anscombe's quartet have (to two decimals) the same means, variances, correlation and regression line. What is the lesson?

Q4

In Anscombe's dataset IV, all x values except one are equal to 8. Why is its correlation coefficient (0.817) misleading?

Q5

What did the Datasaurus Dozen (Matejka & Fitzmaurice, 2017) demonstrate?

Q6

Why can a person find one red dot among hundreds of blue dots almost instantly?

Q7

"External cognition" in visualization means:

Q8

In the reference model of Card, Mackinlay and Shneiderman, which step chooses marks and channels for the data attributes?

Q9

Which is Shneiderman's visual information-seeking mantra?

Q10

John Snow's 1854 map of cholera deaths in London is famous because it:

Q11

Which person is credited with inventing the line chart, the bar chart and the pie chart (1786–1801)?

Q12

Altair is built on which visualization grammar?

Q13

What does the suffix :T mean in the Altair encoding x="date:T"?

Q14

Which Altair encoding computes the average PM2.5 for each value on the x axis?

Q15

In Altair 5, what is alt.Chart(df).mark_line().to_dict()["mark"]?

Q16

A pandas DataFrame has a datetime column d, an integer column v and a string column c. With alt.Chart(df).mark_point().encode(x="d", y="v", color="c") (no type suffixes), which types does Altair infer?

Q17

Why is a declarative chart specification (e.g., Vega-Lite JSON) especially useful when AI generates charts?

Q18

An AI assistant plotted "average trip duration by month" but sorted the months as April, August, December, February… What most likely went wrong?

Q19

An AI tool computed "average PM2.5 per city" by averaging the daily averages of districts that have different numbers of sensors. What is the risk?

Q20

Which of the following is the best use of an AI assistant in this course?

Q21

Which tool family is the best fit for a city official who wants to drag and drop fields into a dashboard without writing code?

Q22

What is the main limitation of D3.js compared with Altair?

Q23

Minard's 1869 map of Napoleon's Russian campaign is admired mainly because it:

Q24

A language model writes "PM2.5 rose sharply in Lucheng in March" as a caption for a chart, but the data shows a small decrease. What kind of AI failure is this?

Q25

Which statement about the CityPulse users is correct for design?

Q26

Florence Nightingale's polar-area diagrams (1858) were used to argue that:

Q27 Short answer

Explain, using Anscombe's quartet, why an analyst who only reads summary statistics (or an AI summary of them) can draw wrong conclusions. Describe what each of the four plots shows.

Q28 Short answer

Describe the visualization pipeline (reference model) for a CityPulse chart "average PM2.5 per day and district for March", naming what happens at each step and where interaction can act.

Q29 Short answer

Write five rules for using AI assistants in a visualization project (what is allowed, what must be verified, what must be disclosed), and give one concrete check for each rule.

Q30 Short answer

For the CityPulse platform, identify three different user groups and, for each, one question they want the data to answer and one reason the same chart would not serve all three.