Defining and applying knowledge conversion processes to a visual analytics system

Knowledge-assisted visualization has been a fast growing field because it directly integrates and utilizes domain knowledge to produce effective data visualization. However, most existing knowledge-assisted visualization applications focus on integrating domain knowledge that is tailored only for specific analytical tasks. This reflects not only the different understandings of what “knowledge” is in visualization, but also the difficulties in generalizing and reapplying knowledge to new problems or domains. In this paper, we differentiate knowledge into two types, tacit and explicit, and suggest four conversion processes between them (internalization, externalization, collaboration, and combination) that could be included in knowledge-assisted visualizations. We demonstrate the applications of these four processes in a bridge visual analytical system for the US Department of Transportation and discuss their roles and utilities in real-life scenarios.