This paper presents Drillboards, a novel approach to designing interactive and adaptive visualization dashboards. Unlike traditional static dashboards, Drillboards employ a hierarchical structure of coordinated charts, enabling users to navigate between different levels of data granularity.
The paper introduces a formal vocabulary of chart representations and a set of rules for merging multiple charts into composite representations. This allows for the creation of a hierarchy where the baseline dashboard, containing all charts at the highest detail, sits at the bottom. Each level up in the hierarchy represents a more abstract and simplified view, achieved by merging charts from the level below.
The authors also present DrillVis, an authoring tool for creating Drillboards. DrillVis allows users to load multidimensional datasets, create a baseline dashboard, and then iteratively build the aggregation hierarchy using the defined merge operations. The tool supports predefined views, enabling authors to tailor the level of detail for different user expertise levels.
The paper demonstrates the utility of DrillVis and the Drillboards concept through a user study involving domain experts and casual end-users. The study revealed that Drillboards effectively communicate data insights to users with varying expertise levels. Experts found the tool valuable for conveying the context and provenance of their data, while novices could quickly grasp the intended message through the hierarchical structure and predefined views.
The paper concludes by discussing the potential of Drillboards in facilitating personalized data exploration and identifies limitations of the current implementation, such as the lack of support for complex network graphs and advanced bespoke visualizations. Future research directions include exploring automated methods for hierarchy generation and investigating the application of Drillboards in different domains.
A otro idioma
del contenido fuente
arxiv.org
Ideas clave extraídas de
by Sungbok Shin... a las arxiv.org 10-17-2024
https://arxiv.org/pdf/2410.12744.pdfConsultas más profundas