Data Visualization for and with AI
Beschrijving
Data visualization is used to communicate statistics and information in a visual manner. This is used frequently in research & engineering, in evidence-based methods (decision and policymaking) and in multiple stages of the development and deployment of data- and AI-driven systems.
This course is situated uniquely at the intersection of data visualization & AI systems. In this course you will learn about the different aspects of data visualization, current best practices, and gain experience in hands-on use of (programming) tools to create data visualizations and dashboards.
Specific attention will be paid to the use of data visualization in development and deployment of AI systems. This includes techniques for interactive exploration of data and improvement of the quality of data and machine learning models. We review methods such as dimensionality reduction, outlier detection, bias/fairness assessment, and XAI methods. Notably, we also survey the converse: (Gen)AI methods to automate the data visualization process itself.
More info? Click here -> https://www.ugain.ugent.be/DVAI2025.htm
Programma
Contents
- Why data visualization: in general & use cases
- Theory and principles of data visualization
- Theory of perception and design mantras
- Visualization of various data types: time series, maps, and graphs
- Dashboards, interactive graphs, and tools to develop web apps
- Visualisation to explore data and AI models, including data quality and bias/fairness
- AI to automate data visualization
- Evaluation of (interactive) visualisation
The course consists of lectures and a few assignments in small groups. These assignments consist of the implementation of (interactive) graphs and dashboards using techniques presented in the lectures, and writing a brief written report with critical reflection. Each assignment is introduced in a tutorial seminar, where the necessary tools are introduced. The assignments partially build on each other. Both the implementation and reflection reports are graded and students will receive feedback on them.
Final competences
- Be able to explain what data visualization is and motivate its use
- Have knowledge of human perception in the context of data visualization
- Have knowledge of and be able to apply the theory of data visualization and design principles
- Appropriately apply existing (semi-automatic) tools for make graphs for different data types and structured data: time series, maps, graphs
- Be able to design and implement interactive graphs and dashboards (as a web app)
- Be familiar with the application of visualization in data exploration and the development of AI models, including exploration and improvement of data quality and bias/fairness
- Critically reflect on (interactive) data visualizations and dashboards and have knowledge on how to evaluate these
