Fundamentals of Data Visualization

Date:

June 25 to 28. MORNING: 9 to 13h (June 25, 26 and 27) and 9 to 12h (June 28).

Instructor

I am Associate Professor at UPC.

I have created and taught numerous courses on Information Visualization, Data Visualization, and Scientific visualization both at Degree and Master level.

My research is mostly on Data Visualization and Scientific Visualization, with large data sets and biomedical models, but I also work on 3D (Virtual Reality) interaction and collaboration, as well as applications of deep learning to visualization.

Language

English

Description

The objective of this course is to teach the fundamental concepts of data visualization. It will not be a course on how to implement a certain chart using a concrete library, but the implementation will be used only to illustrate the theoretical basis.

Course goals

Computer-based visualization systems provide visual representations of datasets intended to help people carry out some task more efficiently. To do so, users must understand the relationship between data, tasks, and users, since all three need to be taken into account when creating visualizations.

The goals of the course are to teach students what types of visualization serve which purposes, what are the limitations of the human perceptual visual system that play a role in visualization, and how to communicate effectively. To implement examples, I will use a Python library, named Altair, but the core part of the course will consist on understanding what makes a visualization effective, what not, and get some background on the different visualization techniques that are available.

The course will also include contents on interaction, the generation of multiple views for problem solving, and techniques for dimensionality reduction used in visualization.

Throughout the course, example visualizations as well as problems will be discussed in class.

Course contents

The contents of the course will visit the following concepts:

Concepts and history. Data, tasks, users.
Common visualization techniques.
Perception in visualization.
Visualization techniques for geospatial data, text, and hierarchical data.
Multiple views and interaction.
Dimensionality reduction techniques for visualization.

Prerequisites

No prerequisite. The implementation can be carried out using any visualization library. I will propose Altair, over Python, but no deep knowledge of the language is required to use it.

Targeted at

Students that need or want to understand how to communicate with data.

Evaluation

An visualization design exercise will be proposed. Students will need to solve it using any library. They can use Altair if they wish, since I will provide a tutorial on the library. But the idea is to leverage the theoretical concepts given throughout the course to achieve effective visualizations, independently on the tool.

Software requirements

If the students want to use some custom visualization software, they have to bring their laptops. Otherwise all the development can be carried out through Google Colab, so only a browser (preferably Google Chrome) is necessary.