Fig. 1. A node-link tree with corresponding treemap and bar chart representations. Treemap design parameters that can affect perception of rectangle area include the aspect ratios of rectangles (top middle, top right), rectangle luminance (bottom left) and border thickness (bottom middle). Bar charts are an alternative encoding of leaf nodes' data that use length rather than area. At lower data densities bar charts can be easier to read, but as the amount of data increases bar charts become less effective because they are not as space-efficient as treemaps. In addition, bar charts do not directly encode the hierarchical structure of a tree.Abstract-Treemaps are space-filling visualizations that make efficient use of limited display space to depict large amounts of hierarchical data. Creating perceptually effective treemaps requires carefully managing a number of design parameters including the aspect ratio and luminance of rectangles. Moreover, treemaps encode values using area, which has been found to be less accurate than judgments of other visual encodings, such as length. We conduct a series of controlled experiments aimed at producing a set of design guidelines for creating effective rectangular treemaps. We find no evidence that luminance affects area judgments, but observe that aspect ratio does have an effect. Specifically, we find that the accuracy of area comparisons suffers when the compared rectangles have extreme aspect ratios or when both are squares. Contrary to common assumptions, the optimal distribution of rectangle aspect ratios within a treemap should include non-squares, but should avoid extreme aspect ratios. We then compare treemaps with hierarchical bar chart displays to identify the data densities at which length-encoded bar charts become less effective than area-encoded treemaps. We report the transition points at which treemaps exhibit judgment accuracy on par with bar charts for both leaf and non-leaf tree nodes. We also find that even at relatively low data densities treemaps result in faster comparisons than bar charts. Based on these results, we present a set of guidelines for the effective use of treemaps.

Poorly designed charts are prevalent in reports, magazines, books and on the Web. Most of these charts are only available as bitmap images; without access to the underlying data it is prohibitively difficult for viewers to create more effective visual representations. In response we present ReVision, a system that automatically redesigns visualizations to improve graphical perception. Given a bitmap image of a chart as input, ReVision applies computer vision and machine learning techniques to identify the chart type (e.g., pie chart, bar chart, scatterplot, etc.). It then extracts the graphical marks and infers the underlying data. Using a corpus of images drawn from the web, ReVision achieves an image classification accuracy of 96% across ten chart categories. It also accurately extracts marks from 79% of bar charts and 62% of pie charts, and from these charts it successfully extracts the data from 71% of bar charts and 64% of pie charts. ReVision then applies perceptually-based design principles to populate an interactive gallery of redesigned charts. With this interface, users can view alternative chart designs and retarget content to different visual styles.

Original chart 1Chart with gridlines 2Line to illustrate trend 3Mean line 4 Fig. 1. In this chart of the European Union's budget by the BBC [7] the original design (1) forces viewers to mentally project a line to the y-axis to extract values. (2) A gridline overlay provides visual anchors, which can simplify the process of extracting values. (3) A line overlay encodes the data redundantly but better illustrates the trends in the data across time. (4) Finally, a statistical summary overlay depicts the mean value of the data so that viewers can easily compare each year's budget to the average budget across the years. All of these overlays were generated by our system without access to the underlying data, based on automatic extraction of the chart's mark and axis properties.Abstract-Reading a visualization can involve a number of tasks such as extracting, comparing or aggregating numerical values. Yet, most of the charts that are published in newspapers, reports, books, and on the Web only support a subset of these tasks. In this paper we introduce graphical overlays-visual elements that are layered onto charts to facilitate a larger set of chart reading tasks. These overlays directly support the lower-level perceptual and cognitive processes that viewers must perform to read a chart. We identify five main types of overlays that support these processes; the overlays can provide (1) reference structures such as gridlines, (2) highlights such as outlines around important marks, (3) redundant encodings such as numerical data labels, (4) summary statistics such as the mean or max and (5) annotations such as descriptive text for context. We then present an automated system that applies user-chosen graphical overlays to existing chart bitmaps. Our approach is based on the insight that generating most of these graphical overlays only requires knowing the properties of the visual marks and axes that encode the data, but does not require access to the underlying data values. Thus, our system analyzes the chart bitmap to extract only the properties necessary to generate the desired overlay. We also discuss techniques for generating interactive overlays that provide additional controls to viewers. We demonstrate several examples of each overlay type for bar, pie and line charts.

News articles, reports, blog posts and academic papers often include graphical charts that serve to visually reinforce arguments presented in the text. To help readers better understand the relation between the text and the chart, we present a crowdsourcing pipeline to extract the references between them. Specifically, we give crowd workers paragraph-chart pairs and ask them to select text phrases as well as the corresponding visual marks in the chart. We then apply automated clustering and merging techniques to unify the references generated by multiple workers into a single set. Comparing the crowdsourced references to a set of gold standard references using a distance measure based on the F 1 score, we find that the average distance between the raw set of references produced by a single worker and the gold standard is 0.54 (out of a max of 1.0). When we apply clustering and merging techniques the average distance between the unified set of references and the gold standard reduces to 0.39; an improvement of 27%. We conclude with an interactive document viewing application that uses the extracted references; readers can select phrases in the text and the system highlights the related marks in the chart.

No abstract