IEEE VIS Publication Dataset

Non-traditional applications of scientific data challenge the typical approaches to visualization. In particular popular scientific visualization strategies fail when the expertise of the data consumer is in a different field than the one that generated the data and data from the user's domain must be utilized as well. This problem occurs when predictive weather simulations are used for a number of weather-sensitive applications. A data fusion approach is adopted for visualization design and utilized for specific example problems.

Multi-dimensional entities are modeled, displayed and understood with a new algorithm vectorizing data of any dimensionality. This algorithm is called SBP; it is a vectorized generalization of parallel coordinates. Classic geometries of any dimensionality can be demonstrated to facilitate perception and understanding of the shapes generated by this algorithm. SBP images of a 4D line, a circle and 3D and 4D spherical helices are shown. A strategy for synthesizing multi-dimensional models matching multi-dimensional data is presented. Current applications include data mining; modeling data-defined structures of scientific interest such as protein structure and Calabi-Yau figures as multi-dimensional geometric entities; generating vector-fused data signature fingerprints of classic frequency spectra that identify substances; and treating complex targets as multi-dimensional entities for automatic target recognition. SBP vector data signatures apply to all pattern recognition problems.

The microscopic analysis of time dependent 3D live cells provides considerable challenges to visualization. Effective visualization can provide insight into the structure and functioning of living cells. The paper presents a case study in which a number of visualization techniques were applied to analyze a specific problem in cell biology: the condensation and de-condensation of chromosomes during cell division. The spatial complexity of the data required sophisticated presentation techniques. The interactive virtual reality enabled visualization system, proteus, specially equipped for time dependent 3D data sets is described. An important feature of proteus is that it is extendible to cope with application-specific demands.

We demonstrate the use of a combination of perceptually effective techniques for visualizing magnetic field data from the DIII-D Tokamak. These techniques can be implemented to run very efficiently on machines with hardware support for OpenGL. Interactive speeds facilitate clear communication of magnetic field structure, enhancing fusion scientists' understanding of their data, and thereby accelerating their research.

One of the main research topics in scientific visualization is to "visualize the appropriate features" of a certain structure or data set. Geodesics are very important in geometry and physics, but there is one major problem which prevents scientists from using them as a visualization tool: the differential equations for geodesics are very complicated and in most cases numerical algorithms must be used. There is always a certain approximation error involved. How can you be sure to visualize the features and not only the approximation quality. The paper presents an algorithm to overcome this problem. It consists of two parts. In the first, a geometric method for the construction of geodesics of arbitrary surfaces is introduced. This method is based on the fundamental property that geodesics are a generalization of straight lines on plains. In the second part these geodesics are used to generate local nets on the surfaces.

Using inductive learning techniques to construct classification models from large, high-dimensional data sets is a useful way to make predictions in complex domains. However, these models can be difficult for users to understand. We have developed a set of visualization methods that help users to understand and analyze the behavior of learned models, including techniques for high-dimensional data space projection, display of probabilistic predictions, variable/class correlation, and instance mapping. We show the results of applying these techniques to models constructed from a benchmark data set of census data, and draw conclusions about the utility of these methods for model understanding.

Visualization techniques enable scientists to interactively explore 3D data sets, segmenting and cutting them to reveal inner structure. While powerful, these techniques suffer from one serious flaw-the images they create are displayed on a flat piece of glass or paper. It is not really 3D-it can only be made to appear 3D. We describe the construction of 3D physical models from volumetric data. Using solid freeform fabrication equipment, these models are built as separate interlocking pieces that express in physical form the segmentation and cutting operations common in display-based visualization.

Accurately and automatically conveying the structure of a volume model is a problem that has not been fully solved by existing volume rendering approaches. Physics-based volume rendering approaches create images which may match the appearance of translucent materials in nature but may not embody important structural details. Transfer function approaches allow flexible design of the volume appearance but generally require substantial hand-tuning for each new data set in order to be effective. We introduce the volume illustration approach, combining the familiarity of a physics-based illumination model with the ability to enhance important features using non-photorealistic rendering techniques. Since the features to be enhanced are defined on the basis of local volume characteristics rather than volume sample values, the application of volume illustration techniques requires less manual tuning than the design of a good transfer function. Volume illustration provides a flexible unified framework for enhancing structural perception of volume models through the amplification of features and the addition of illumination effects.

WEAVE (Workbench Environment for Analysis and Visual Exploration) is an environment for creating interactive visualization applications. WEAVE differs from previous systems in that it provides transparent linking between custom 3D visualizations and multidimensional statistical representations, and provides interactive color brushing between all visualizations. The authors demonstrate how WEAVE can be used to rapidly prototype a biomedical application, weaving together simulation data, measurement data, and 3D anatomical data concerning the propagation of excitation in the heart. These linked statistical and custom three-dimensional visualizations of the heart can allow scientists to more effectively study the correspondence of structure and behavior.

This paper proposes a new technique to visualize dependencies among cells in a spreadsheet. In this way, the system firstly visualizes a spreadsheet on a plane in three-dimensional space, and draws arcs between interrelated cells. By allowing a user to select an arbitrary cell and lift it up with direct manipulation, the system utilizes the third dimension to ameliorate visual occlusion of crossing arcs. As the user lifts a focused cell up, the interrelated cells are lifted up together; thus hidden dataflow networks can be visually intelligible interactively. Because spreadsheets are aimed at calculation itself rather than appearances of outputs, their mechanism is relatively invisible and not obvious for ordinary users. Our visualization helps such users to understand structures and mechanism of spreadsheets

This paper aims to give a systematic account of focus+context visualization techniques, i.e. visualizations which aim to give users integrated visual access to details and context in a data set. We introduce the notion that there are different orders of information visualization with focus+context being a second-order visualization and provide a formal framework for describing and constructing focus+context visualizations

Many real-world KDD (Knowledge Discovery & Data Mining) applications involve the navigation of large volumes of information on the web, such as, Internet resources, hot topics, and telecom phone switches. Quite often users feel lost, confused and overwhelmed with displays that contain too much information. This paper discusses a new content-driven visual mining infrastructure called VisMine, that uses several innovative techniques: (1) hidden visual structure and relationships for uncluttering displays; (2) simultaneous visual presentations for high-dimensional knowledge discovery; and (3) a new visual interface to plug in existing graphic toolkits for expanding its use in a wide variety of visual applications. We have applied this infrastructure to three data mining visualization applications-topic hierarchy for document navigation, web-based trouble shooting, and telecom switch mining

We have developed a technique, Aggregate Towers, that allows geospatial data to be visualized across a range of map scales. We use a combination of data aggregation algorithms and dynamically aggregating data markers (e.g., icons or symbols) to accommodate interactive zooming by a user while maintaining a representation that remains intuitive, consistent across multiple scales and uncluttered. This approach implicitly generates multiple levels of overview displays from a single set of underlying data

A new method is presented to get an insight into univariate time series data. The problem addressed is how to identify patterns and trends on multiple time scales (days, weeks, seasons) simultaneously. The solution presented is to cluster similar daily data patterns, and to visualize the average patterns as graphs and the corresponding days on a calendar. This presentation provides a quick insight into both standard and exceptional patterns. Furthermore, it is well suited to interactive exploration. Two applications, numbers of employees present and energy consumption, are presented

Constellation is a visualization system for the results of queries from the MindNet natural language semantic network. Constellation is targeted at helping MindNet's creators and users refine their algorithms, as opposed to understanding the structure of language. We designed a special-purpose graph layout algorithm which exploits higher-level structure in addition to the basic node and edge connectivity. Our layout prioritizes the creation of a semantic space to encode plausibility instead of traditional graph drawing metrics like minimizing edge crossings. We make careful use of several perceptual channels both to minimize the visual impact of edge crossings and to emphasize highlighted constellations of nodes and edges

A new method is presented for the visualization of hierarchical information, such as directory structures and organization structures. Cushion treemaps inherit the elegance of standard treemaps: compact, space-filling displays of hierarchical information, based on recursive subdivision of a rectangular image space. Intuitive shading is used to provide insight in the hierarchical structure. During the subdivision, ridges are added per rectangle, which are rendered with a simple shading model. The result is a surface that consists of recursive cushions. The method is efficient, effective, easy to use and implement, and has a wide applicability

We examine how animating a viewpoint change in a spatial information system affects a user's ability to build a mental map of the information in the space. We found that animation improves users' ability to reconstruct the information space, with no penalty on task performance time. We believe that this study provides strong evidence for adding animated transitions in many applications with fixed spatial data where the user navigates around the data space

Domain Analysis for Data Visualization (DADV) is a technique to use when investigating a domain where data visualizations are going to be designed and added to existing software systems. DADV was used to design the data visualization in VisEIO-LCA, which is a framework to visualize environmental data about products. Most of the visualizations are designed using the following stages: formatting data in tables, selecting visual structures, and rendering the data on the screen. Although many visualization authors perform implicit domain analysis, in this paper domain analysis is added explicitly to the process of designing visualizations with the goal of producing move usable software tools. Environmental Life-Cycle Assessment (LCA) is used as a test bed for this technique

This paper describes concepts that underlie the design and implementation of an information exploration system that allows users to impose arbitrary hierarchical organizations on their data. Such hierarchies allow a user to embed important semantic information into the hierarchy definition. Our goal is to recognize the significance of this implicit information and to utilize it in the hierarchy visualization. The innovative features of our system include the dynamic modification of the hierarchy definitions and the definition and implementation of a set of layout algorithms that utilize semantic information implicit in the tree construction

Dynamic queries offer continuous feedback during range queries, and have been shown to be effective and satisfying. Recent work has extended them to datasets of 100,000 objects and, separately, to queries involving relations among multiple objects. The latter work enables filtering houses by properties of their owners, for instance. Our primary concern is providing feedback from histograms during dynamic query. The height of each histogram bar shows the count of selected objects whose attribute value falls into a given range. Unfortunately, previous efficient algorithms for single object queries overcount in the case of multiple objects if for instance, a house has multiple owners. This paper presents an efficient algorithm that with high probability closely approximates the true counts