Duration
25h Th, 20h Pr, 25h Proj.
Number of credits
Lecturer
Language(s) of instruction
English language
Organisation and examination
Teaching in the second semester
Schedule
Units courses prerequisite and corequisite
Prerequisite or corequisite units are presented within each program
Learning unit contents
Motivation
Data visualization is an important step preceding advanced analyses or modeling. It helps users form an initial intuition about data quality, the potential results of an analysis, the tools to use for more detailed visualization, and the performance of a predictive model. However, creating a relevant and impactful visualization requires appropriate tools and adherence to best-practice guidelines, which can be a genuine source of difficulty for novice users.
In this context, the aim of this course is to enable students to acquire the knowledge and technical skills needed to:
- Explore data using various visualization tools.
- Select and implement relevant tools and appropriate parameters to visualize data and communicate effectively through those visualizations.
- Understand the cognitive aspects involved in interpreting visualizations.
- Think critically about scientific and media visualizations.
Table of contents
1. Introduction to data visualization
a. Motivation, the role of visualization in the scientific process and in communication in general, and good and bad introductory examples
b. How to read and analyze a visualization: what message to convey, and to which audience
c. Data types and their structures
2. Visualization and connections with various fields
a. Cognitive science: mechanisms of human perception, Gestalt principles, and human perceptual biases (magnitudes, areas, colors, etc.)
b. Visualization as an interface for human-computer interaction
c. Elements of computer graphics
d. Visualization, particularly in data science
3. Elements of color theory
a. Color spaces, mixing, and contrasts
b. Accessibility, color blindness, constraints, and human interpretation
4. Visualizations of "raw data"
a. Univariate categorical and numerical data: tools such as bar plots, histograms, and box plots
b. Bivariate data: tools such as scatter plots and 2D histograms
c. Multivariate data (using color, size, marker shape, etc. to represent additional dimensions) and the challenges of high dimensionality
5. Visualization of "functions"
a. Time series (1D->1D)
b. Parametric curves and trajectories (1D->2D/3D)
c. Monochromatic "images," heatmaps, and spatial data (2D->1D)
d. Volumetric data (3D->1D)
e. Displacement fields and flows (2D->2D)
6. Visualization of "structured data"
a. Trees and dendrograms
b. Graphs and networks
c. Structuring and visualization of textual data
7. Practical approach and precautions for use
a. Selecting the right tools for raw data
b. Elements of visualization ethics (manipulation issues, missing data, outliers, duplicates, etc.)
c. Critical analysis of visualizations
8. Dynamic visualization
a. Interactivity in a visualization and sensitivity of the visualization to parameter choices
b. Animation of a visualization
9. Visualization of predictive model outputs
a. Uncertainty and error bars
b. Predictions vs. oracle, coloring large matrices, model "performance," and precautions for interpretation
10. Production of scientific figures
a. Components of a good scientific figure
b. Best-practice rules for scientific visualization, scientific ethics, fidelity of representation and interpretation, reproducibility, and publication constraints
c. Storytelling with a sequence of figures and visual consistency
Learning outcomes of the learning unit
The aim of this course is to enable students to acquire the knowledge and technical skills needed to:
- Explore data using various visualization tools.
- Select and implement relevant tools and appropriate parameters to visualize data and communicate effectively through those visualizations.
- Understand the cognitive aspects involved in interpreting visualizations.
- Think critically about scientific and media visualizations.
Prerequisite knowledge and skills
The prerequisites are covered by compulsory courses (statistics and introduction to programming).
Planned learning activities and teaching methods
Lectures + practical sessions in Python.
Mode of delivery (face to face, distance learning, hybrid learning)
Face-to-face course
Course materials and recommended or required readings
Platform(s) used for course materials:
- eCampus
- MyULiège
Exam(s) in session
Any session
- In-person
written exam ( open-ended questions )
Written work / report
Out-of-session test(s)
Further information:
The assessment consists of two components, each worth 50% of the final grade.
A written examination assessing students' theoretical knowledge, analytical and critical-thinking skills when presented with examples of poor visualizations, and their ability to propose appropriate corrections.
An "individual project" assessing students' ability to produce relevant visualizations and impactful scientific figures from provided data, including an oral defense of their choices and an interpretation of the results obtained.
Important note: The "individual project" may be divided into two or three parts over the course of the semester to avoid an excessive workload at the end of the term. It will primarily involve applying the concepts covered in class. If enrollment is high, group work may be considered. The oral defense may be replaced with a brief written report. The exact arrangements will be determined at the beginning of the semester.
Work placement(s)
Organisational remarks and main changes to the course
Contacts
Adrien Deliege
adrien.deliege@uliege.be