Semantic Interaction: Coupling Cognition and Computation through Usable Interactive Analytics

Abstract

The success of visual analytics is predicated on the ability of users to interactively explore information. Humans think about their data through interactive visual exploration, including testing hypotheses, exploring anomalies, and other cognitive processes of building understanding from data. The claim that these insights are generated as a result of the interaction led the attendees at the Pacific Northwest National Laboratory (PNNL) workshop on 'Semantic Interaction: Coupling Cognition and Computation through Usable Interactive Analytics' to posit that user interaction μst play a more central role in visual analytics systems, serving as the method for coupling cognition and computation. The claims and design principles discussed in this workshop report present research directions to advance visual analytics via a user interaction approach called semantic interaction.