Data-driven Evaluation of Visual Quality Measures

Data-driven Evaluation of Visual Quality Measures

Abstract

Visual quality measures seek to algorithmically imitate human judgments of patterns such as class separability, correlation, or outliers. In this paper, we propose a novel data-driven framework for evaluating such measures. The basic idea is to take a large set of visually encoded data, such as scatterplots, with reliable human 1Cground truth 1D judgements, and to use this human-labeled data to learn how well a measure would predict human judgements on previously unseen data. Measures can then be evaluated based on predictive performance 14an approach that is crucial for generalizing across datasets but has gained little attention so far. To illustrate our framework, we use it to evaluate 15 state-of-the-art class separation measures, using human ground truth data from 828 class separation judgments on color-coded 2D scatterplots

Grafik Top
Authors
  • Sedlmair, Michael
  • Aupetit, Michaël
Grafik Top
Supplemental Material
Shortfacts
Category
Journal Paper
Divisions
Visualization and Data Analysis
Subjects
Computergraphik
Journal or Publication Title
Computer Graphics Forum: the international journal of the Eurographics Association 2015
ISSN
0167-7055
Date
May 2015
Export
Grafik Top