Visualization has become a valuable means for data exploration and analysis. Interactive visualization combines expressive graphical representations and effective user interaction. Although interaction is an important component of visualization approaches, much of the visualization literature tends to pay more attention to the graphical representation than to interaction. The goal of this work is to strengthen the interaction side of visualization. Based on a brief review of general aspects of interaction, we develop an interaction-oriented view on visualization. This view comprises five key aspects: the data, the tasks, the technology, the human, as well as the implementation. Picking up these aspects individually, we elaborate several interaction methods for visualization. We introduce a multi-threading architecture for efficient interactive exploration. We present interaction techniques for different types of data e.g., multivariate data, spatio-temporal data, graphs) and different visualization tasks (e.g., exploratory navigation, visual comparison, visual editing). With respect to technology, we illustrate approaches that utilize modern interaction modalities (e.g., touch, tangibles, proxemics) as well as classic ones. While the human is important throughout this work, we also consider automatic methods to assist the interactive part. In addition to solutions for individual problems, a major contribution of this work is the overarching view of interaction in visualization as a whole. This includes a critical discussion of interaction, the identification of links between the key aspects of interaction, and the formulation of research topics for future work with a focus on interaction.

Autorentext

Christian Tominski received a diploma and doctoral degree from the University of Rostock, Germany, in 2002 and 2006, respectively. He is a senior researcher and lecturer at the Institute for Computer Science at the University of Rostock. His research interests are in visualization and visual analytics. He is particularly interested in the role of interaction for visual data exploration and analysis. He worked on utilizing novel display and interaction devices for interactive visualization and on integrating automatic methods to assist the visualization. Christian has authored and co-authored more than 50 academic publications on new visualization approaches and interaction methods. He also developed a number of visualization tools for time-oriented data, spatio-temporal data, movement data, and graph data. More about Christian's research and demos of visualization prototypes can be found on his website.