Generating visualization-based analysis scenarios from maintenance task descriptions

Abstract

Software visualization is an efficient and flexible tool to inspect and analyze software data at various levels of detail. However, software analysts typically do not have a sufficient background in visualization and cognitive science to select efficient representations and parameters without the help of visualization experts. To overcome this problem, we propose an approach to generate software analysis tasks that use visualization. To this end, we use taxonomies of low-level analytic tasks, high-level interactive tasks, and perceptual rules to design an assistant that proposes analysis scenarios.