Visual Analytics and Uncertainty: Its Not About the Data

Abstract

Uncertainty visualization research has a long history, with contributions from scientific, information, geo-graphic and other visualization perspectives as well as from cognitive and HCI perspectives. But we still do not have generally accepted strategies for leveraging visualization to cope with uncertainty. Here, I argue that taking a visual analytics rather than visualization perspective can overcome this inertia. While uncertainty visualization research has focused on visually signifying and interacting with data uncertainty, taking a visual analytics approach recognizes that the challenge is about much more than uncertain data. The larger challenge is to enable reasoning under uncertainty (in all its forms). In this short paper, I sketch elements of what we know and outline some key challenges for developing visual analytics methods and tools that enable users to cope with uncertainty throughout the processes of sensemaking, decision-making, and action-taking.