Visual Parameter Space Analysis: A Conceptual Framework

Abstract

Various case studies in different application domains have shown the great potential of visual parameter space analysis to support validating and using simulation models. In order to guide and systematize research endeavors in this area, we provide a conceptual framework for visual parameter space analysis problems. The framework is based on our own experience and a structured analysis of the visualization literature. It contains three major components: (1) a data flow model that helps to abstractly describe visual parameter space analysis problems independent of their application domain; (2) a set of four navigation strategies of how parameter space analysis can be supported by visualization tools; and (3) a characterization of six analysis tasks. Based on our framework, we analyze and classify the current body of literature, and identify three open research gaps in visual parameter space analysis. The framework and its discussion are meant to support visualization designers and researchers in characterizing parameter space analysis problems and to guide their design and evaluation processes.

Visualization System Requirements for Data Processing Pipeline Design and Optimization

“…Nevertheless, we note that the basic calculation step, the computation of derived data, and the analysis of output sensitivity feature in both works. Therefore, we see our work as complementary extension of the framework provided by Sedlmair et al [2].…”

“…Our paper takes a different perspective on visual parameter analysis than that theoretically summarized by Sedlmair et al [2] or analyzed in previous scenarios in the literature (see Table 1). Sedlmair et al focus on the tasks performed during data modeling and analyze types of parameters, inputs and outputs of modeling.…”

“…Our research builds on Sedlmair et al's conceptual framework [2], which characterizes data flows, navigation strategies and analysis tasks that take place during visual parameter space analysis. Sedlmair et al's analysis shows that most users adopt a global-to-local navigation strategy that is consistent with Shneiderman's well-known mantra [20].…”

“…Sedlmair et al [2] identify six user tasks: optimization, model output partitioning, model fitting, finding outliers, assessment of output uncertainty, and model sensitivity. Outliers are one aspect of the veracity of a pipeline's inputs, and uncertainty and sensitivity are included as distinct factors in our characterization of users' requirements (see Sec.…”

“…A systematic assessment is needed to show how current visualization functionality matches user's requirements for pipeline design and optimization, and which new functionality still needs to be developed. Previous research has made steps in this direction (e.g., Sedlmair et al's conceptual framework [2]), but the focus was on parameter space analysis rather than pipeline design and optimization.…”