Guidance in Visual Analytics

Theoretical (Analytical):

Practical (Implementation):

Literature Work:

Overview and Problem Statement

Visual analytics is concerned with integrating machine learning and automated data analysis with human intuition in order to effectively solve complex analyis tasks. Guidance in visual analytics describes a mixed-initiative process by which systems aim to provide suggestions to resolve knowledge gaps that user encounter, while users highlight areas of interest and contribute domain knowledge. The resulting mixed-initiative process can be characterized in terms of learning and teaching, where both the system and the user learn from each other while teaching at the same time. 

In this project, you will explore how to guide users through visual text analytics. The provided guidance should be both adaptive in terms of content and visual presentation, learning the users preferences over time and making more and more appropriate suggestions over time. A successful project will also consider how to avoid pitfalls like confirmation bias.


  • Review current approaches to user modeling and adaptive guidance in visual analytics
  • Develop a framework that learns and visualizes user preference
  • Evaluate user performance improvements versus a traditional, non-adaptive system


  • Interest in evaluation of XAI
  • Previous knowledge of (cognitive/perceptual/...) biases is a plus, but not required


  • Scope: Bachelor/Master
  • 6 Month Project, 3 Month Thesis (Bachelor) / 6 Month Thesis (Master)
  • Start: immediately