Output details
11 - Computer Science and Informatics
University of Kent
Assessing the Effect of Visualizations on Bayesian Reasoning through Crowdsourcing
<20> This paper addresses representations for classic Bayesian problems. It assesses six visualizations, four of which are novel, and two textual representations. The study suggests that merely adding visualizations to classic textual problems is not helpful but visualisations are effective when quantitative values are omitted. One visualization is demonstrated in a video http://www.youtube.com/watch?v=D8VZqxcu0I0 that has over 8000 views. The work was presented at IEEE InfoVis, the most prestigious conference in Information Visualization (acceptance rate under 25%), and was one of three papers to receive an honourable mention. As a full paper, it was published in a special issue of TVCG.