Quality Metrics for Visual Analytics of High-Dimensional Data

Visual Analytics of high-dimensional data is particularly challenging. In my talk I will present and discuss quality metrics which have been proposed to help in the visual exploration of patterns in high-dimensional data. In a number of recent approaches, quality metrics have been used to automate the demanding search through large spaces of alternative visualizations, allowing the user to concentrate on the most promising visualizations as suggested by the quality metrics. I will present a set of factors for discriminating the quality metrics, visualization techniques, and the process itself. The process can be described through a reworked version of the well-known information visualization pipeline. The usefulness of the model will be shown by applying it to several visual analytics approaches that are based on quality metrics.