Jump to main content

Quicklinks

Lecture and Exercises Visual Analysis

Lecturer: Prof. Tatiana von Landesberger

Course number: 14722.5007

Dates: Th. 10:00-13:30, 136 Hörsaal XXX (Alte Botanik)

Locations and times of the exercises will be announced in the first lecture.

Teaching Language: English

Contents

The lecture on “Visual Analytics” focuses on the interactive visual analysis of large, complex datasets, combining visualization, human-centered interaction, and automated data analysis (ML, AI) techniques. The lecture shows how analytical reasoning can be supported through interactive visual interfaces, particularly in contexts where the volume, variety, and uncertainty of data challenge traditional analysis methods. Lecture emphasizes how interactive visualization can be supported by ML/AI Methods and how ML/AI Methods can be supported by interactive visualization.

The course covers selected topics in visualization design, interaction techniques, human perception, and data analytics (AI4VIS, VIS4AI), emphasizing their integration to solve application-oriented problems. Students will learn fundamental methods, practical examples, and current research directions in Visual Analytics.

A significant part of the course explores the role of artificial intelligence (AI) and large language models (LLMs) in modern visual analysis workflows. This includes the use of:

  • Dashboards for evaluating and monitoring machine learning models
  • Use of AI and LLM for generating visualizations and dashboards
  • Comparative views of model predictions versus observed data
  • Uncertainty visualization to communicate model confidence and limitations

These applications span various domains, including finance, economics, geosciences, meteorology, medicine, biology, transportation, and sports, where complex data and predictive models are becoming more prevalent. This course prepares students to critically explore, interpret, and communicate insights from advanced, data-driven systems using visual-analytical approaches.
 

Exercises

To accompany the lecture, there are hands-on exercises that provide students with the opportunity to apply their theoretical knowledge to a practical context. Working in small groups, participants design and implement their own visual analytics projects, working through all stages, from data selection and pre-processing to interactive visual representation and interpretation.


These projects allow students to explore real-world datasets, apply the methods discussed in the lecture, and experiment with integrating analytical techniques, interactive visualizations, and, where applicable, AI or machine learning components. Particular emphasis is placed on combining multiple views, linking visual components, and incorporating dynamic exploration to support complex analytical tasks.


The exercises finish in project presentations where students demonstrate their systems, explain their design decisions, and reflect on their analytical outcomes. Through this applied format, the course fosters technical proficiency and critical understanding of the visual analytics process.