TU Wien:Informationsvisualisierung VO (Matkovic)/Exam 2023-09-28
Group A[Bearbeiten | Quelltext bearbeiten]
Question 1[Bearbeiten | Quelltext bearbeiten]
Given a violin plot with some embedded boxplot-like chart, read out the values (or, if not possible, tick “It is not possible to extract this information").
- The maximum is … □ not possible to extract this information
- The mean is … □ not possible to extract this information
- The median is … □ not possible to extract this information
- The interquartile range is … □ not possible to extract this information
Select the correct answer:
The distribution is
- unimodular
- bimodular
- multimodular
- not possible to extract this information
Question 2[Bearbeiten | Quelltext bearbeiten]
Given was an adjacency matrix of an undirected graph.
| A | B | C | D | E | |
|---|---|---|---|---|---|
| A | 0 | 1 | 1 | 0 | 0 |
| B | 1 | 0 | 0 | 1 | 0 |
| C | 1 | 0 | 0 | 1 | 1 |
| D | 0 | 1 | 1 | 0 | 0 |
| E | 0 | 0 | 1 | 0 | 0 |
Draw a node-link diagram of the resulting graph.
Question 3 (Open Question)[Bearbeiten | Quelltext bearbeiten]
Assume you have a (tabular) dataset that contains data of about 100 companies. There are 64 numerical attributes, such as
- Net. revenue
- Liabilities
- Number of employees
- Sales
- …
Suggest a visualization that allows to detect companies which have a lot in common and that allows to detect outliers. Sketch the visualization (just a raw sketch to give an idea, nothing fancy) and additionally comment what you would plot if the sketch is not sufficient. Also mention what sort of computations you would need beforehand.
Question 4 (Multiple Choice)[Bearbeiten | Quelltext bearbeiten]
Which statements about a kernel density estimation are correct? A KDE is
- non-parametric
- data must be normally distributed
- the kernel width can be varied
- (two more options I cannot recall which were false)