Trustworthy Data Visualization

“Without good data we are flying blind if you can’t see it you can’t solve it.” Kofi Anan, Nature, 2018.

“I think what has changed permanently, ..., is the notion that data, data visualization are useful to ministers during the crisis. Actually they were pulling.” Patrick Vallance, AI UK, April 2022.

One of the lessons of the COVID19 pandemic in the UK has been the demand from decision makers for the daily visualization of data about the state of the country. The decision makers included ministers, politicians and the public, each needing to understand the current situation to make decisions for the country and in their everyday lives.

This demand has exposed some important questions for visualization researchers:

• How to ensure data visualizations are trustworthy?
• How to validate the quality of data visualizations?
• How to engage people in data visualizations?

We will demonstrate how our recent research applying information theoretic methods to the challenges of urban data visualization addresses these questions and argue that successful data visualization increasingly needs rigorous underpinning science.

Speaker
Professor Nick Holliman researches the science and engineering of visualization and visual analytics, addressing the fundamental challenges of visualization in successful human-machine teaming. He has demonstrated how novel visualization methods work in practice in scalable cloud-based software tools and award-winning 3D visualizations. Nick has experience delivering global commercial impact for research outputs from both industrial and academic laboratories. He is Director of CUSP London and leads the King’s Turing Network Development project with the Alan Turing Institute, where he co-founded and co-convenes the Visualization Interest Group.