Maureen C. Stone

iStuff: a physical user interface toolkit for ubiquitous computing environments

The iStuff toolkit of physical devices, and the flexible software infrastructure to support it, were designed to simplify the exploration of novel interaction techniques in the post-desktop era of multiple users, devices, systems and applications collaborating in an interactive environment. The toolkit leverages an existing interactive workspace in-frastructure, making it lightweight and platform independent. The supporting software framework includes a dynamically configurable intermediary to simplify the mapping of devices to applications. We describe the iStuff architecture and provide several examples of iStuff, organized into a design space of ubiquitous computing interaction components. The main contribution is a physical toolkit for distributed, heterogeneous environments with run-time retargetable device data flow. We conclude with some insights and experiences derived from using this toolkit and framework to prototype experimental interaction techniques for ubiquitous computing environments.

Selecting semantically-resonant colors for data visualization

We introduce an algorithm for automatic selection of semantically‐resonant colors to represent data (e.g., using blue for data about “oceans”, or pink for “love”). Given a set of categorical values and a target color palette, our algorithm matches each data value with a unique color. Values are mapped to colors by collecting representative images, analyzing image color distributions to determine value‐color affinity scores, and choosing an optimal assignment. Our affinity score balances the probability of a color with how well it discriminates among data values. A controlled study shows that expert‐chosen semantically‐resonant colors improve speed on chart reading tasks compared to a standard palette, and that our algorithm selects colors that lead to similar gains. A second study verifies that our algorithm effectively selects colors across a variety of data categories.

Color naming models for color selection, image editing and palette design

Our ability to reliably name colors provides a link between visual perception and symbolic cognition. In this paper, we investigate how a statistical model of color naming can enable user interfaces to meaningfully mimic this link and support novel interactions. We present a method for constructing a probabilistic model of color naming from a large, unconstrained set of human color name judgments. We describe how the model can be used to map between colors and names and define metrics for color saliency (how reliably a color is named) and color name distance (the similarity between colors based on naming patterns). We then present a series of applications that demonstrate how color naming models can enhance graphical interfaces: a color dictionary & thesaurus, name-based pixel selection methods for image editing, and evaluation aids for color palette design.

In Color Perception, Size Matters

In designing colors for digital visualization systems. one of the most critical factors is the interaction between size and color appearance. Whereas artists and visual designers understand this, recommendations for using color in visualization rarely address the issue, or even recognize it. The goal is to keep the user focused on the data, with color simply one of many tools used toward problem solving and discovery.

A Linguistic Approach to Categorical Color Assignment for Data Visualization

When data categories have strong color associations, it is useful to use these semantically meaningful concept-color associations in data visualizations. In this paper, we explore how linguistic information about the terms defining the data can be used to generate semantically meaningful colors. To do this effectively, we need first to establish that a term has a strong semantic color association, then discover which color or colors express it. Using co-occurrence measures of color name frequencies from Google n-grams, we define a measure for colorability that describes how strongly associated a given term is to any of a set of basic color terms. We then show how this colorability score can be used with additional semantic analysis to rank and retrieve a representative color from Google Images. Alternatively, we use symbolic relationships defined by WordNet to select identity colors for categories such as countries or brands. To create visually distinct color palettes, we use k-means clustering to create visually distinct sets, iteratively reassigning terms with multiple basic color associations as needed. This can be additionally constrained to use colors only in a predefined palette.

The Effect of Colour and Transparency on the Perception of Overlaid Grids

Overlaid reference elements need to be sufficiently visible to effectively relate to the underlying information, but not so obtrusive that they clutter the presentation. We seek to create guidelines for presenting such structures through experimental studies to define boundary conditions for visual intrusiveness. We base our work on the practice of designers, who use transparency to integrate overlaid grids with their underlying imagery. Previous work discovered a useful range of alpha values for black or white grids overlayed on scatterplot images rendered in shades of gray over gray backgrounds of different lightness values. This work compares black grids to blue and red ones on different image types of scatterplots and maps. We expected that the coloured grids over grayscale images would be more visually salient than black ones, resulting in lower alpha values. Instead, we found that there was no significant difference between the boundaries set for red and black grids, but that the boundaries for blue grids were set consistently higher (more opaque). As in our previous study, alpha values are affected by image density rather than image type, and are consistently lower than many default settings. These results have implications for the design of subtle reference structures.

Whisper, Don't Scream: Grids and Transparency

Visual elements such as grids, labels, and contour lines act as reference structures that support the primary information being presented. Such structures need to be usefully visible, but not so obtrusive that they clutter the presentation. Visual designers know how to carefully manage transparency and layering in an image to balance these elements. We want the presentation of these structures in complex, dynamic, computer-generated visualizations to reflect the same subtlety and comfort of good design. Our goal is to determine the physical, perceptual, and cognitive characteristics of such structures in a way that enables automatic presentation. Our approach to this problem does not try to characterize ideal” or best,” but instead seeks boundary conditions that define a range of visible yet subtle legibility. All presentations that are clearly bad lie outside of this range, and can easily be avoided. In this paper, we report three experiments investigating the effects of grid color and spacing on these boundary conditions, defined by manipulating the transparency (alpha) of thin rectangular grids over scatter plots. Our results show that while there is some variation due to user preference and image properties, bounding alpha allows us to reliably predict a range of usable yet unobtrusive grids over a wide variety of conditions.

Affective Color in Visualization

Communicating the right affect, a feeling, experience or emotion, is critical in creating engaging visual communication. We carried out three studies examining how different color properties (lightness, chroma and hue) and different palette properties (combinations and distribution of colors) contribute to different affective interpretations in information visualization where the numbers of colors is typically smaller than the rich palettes used in design. Our results show how color and palette properties can be manipulated to achieve affective expressiveness even in the small sets of colors used for data encoding in information visualization.

A Focus + Context Technique for Visualizing a Document Collection

Investigative analysts need overviews of large amounts of data, which is a challenge when working with non-numerical data such as document collections. We present Semantic Zoom View (SZV), an interactive document collection visualization implemented as part of the CZSaw visual analytics system. SZV uses a focus + context technique to provide an overview with details on demand through interactive semantic zooming. SZV lets an analyst easily and quickly see the main topics of a document collection while keeping surrounding documents visible for context. Working within a single integrated visualization, an analyst can also quickly find related documents and break a large document collection into smaller meaningful groups. SZVs focus + context technique was compared to an overview + detail version for finding answers within a document collection and results indicated its strength for maintaining visibility of a full overview when document contents are accessed.

Model based interactive analysis of interwoven, imprecise narratives: VAST 2010 mini challenge 1 award: Outstanding interaction model

CZSaw [1] is a visual analytics tool for sense-making across entities, documents, and relations with a focus on supporting the analysis process. It uses a variety of flexible data visualizations to represent and explore networks of entities and relations from different perspectives. CZSaw supports clustering documents and entities into smaller groups to make sense of them and weave individual facts into a complete picture. CZSaw also provides entity refinement functions to support interactive data cleaning. Its dependency propagation mechanism speeds the analysis sense-making loop by automatically synchronizing data and views, and propagating changes to the whole system.