Towards User-Adaptive Visualizations
By Cristina Conati
In this talk, I will describe research aimed at devising user-adaptive visualizations, i.e.. intelligent visualization techniques that can automatically adapt to the needs of individual users as they interact with the visualization software.
I will start by describing the results of a user study we ran to investigate the effect of various user cognitive abilities on the effectiveness of two alternative data visualization techniques for representing value change. We found that some of the tested cognitive abilities can either predict which of the two visualizations is more effective for a given user, or predict user's performance with a specific visualization.
I will then discuss the implications of these results for devising user-adaptive visualizations, as well as some of the main challenges in this research.