Code-Space Quality Evaluation for Information Visualization

Abstract

The quality evaluation is essential to creating effective data visualization designs. The data visualization research community has produced many quality metrics for evaluating data visualization. However, these quality metrics are rarely integrated into popular data visualization tools. As a result, most data visualization creators are either not aware of these quality metrics or do not know how to apply these metrics to the visualization creation process. In this paper, we propose a novel quality evaluation method that integrates quality metrics into popular data visualization programming tools. Our main contribution is a code-space quality evaluation method, different from the traditional image-space or data-space quality evaluation method. Using our method, a visualization programmer passes a coded data visualization design to a quality evaluation function that generates warnings, comments, and design recommendations. This allows users to integrate quality checks into the design process.