Hacking and Debugging the User in Visual Analytics

Abstract

In visual analytics, it is generally believed that "interaction is the analysis," and that a user's interactions are a reflection of the user's reasoning and cognitive processes during data analysis and exploration. It is through interaction that humans leverage their curiosity, intuition, and creativity to discover patterns, relationships, and other phenomena within data. In order to create effective visual analytics systems, it is essential that we understand how a user's interactions impact their analysis.

In this talk, I will present techniques that we have developed to analyze the user's interactions in order to better understand the user's reasoning and cognitive processes. Using machine learning methods, we demonstrate that we can leverage the user's interactions to: (a) model the user and learn the user's analysis behaviors, and (b) model the data to learn parameters of a distance metric that best represent high-dimensional data. In addition, using a cognitive technique known as "cognitive priming," we demonstrate that we can subtly but consistently influence the user's analysis behaviors in a predictable manner.

Bio: Remco Chang is an Assistant Professor in the Computer Science Department at Tufts University. He received his B.S from Johns Hopkins University in 1997 in Computer Science and Economics, MSc from Brown University in 2000, and Ph.D. in computer science from UNC Charlotte in 2009. Prior to his Ph.D., he worked for Boeing developing real-time flight tracking and visualization software, followed by a position at UNC Charlotte as a research scientist. His current research interests include visual analytics, information visualization, and HCI. His research has been funded by NSF, DHS, and Draper.