IEEE VIS Publication Dataset

Time series are an important type of data with applications in virtually every aspect of the real world. Often a large number of time series have to be monitored and analyzed in parallel. Sets of time series may show intrinsic hierarchical relationships and varying degrees of importance among the individual time series. Effective techniques for visually analyzing large sets of time series should encode the relative importance and hierarchical ordering of the time series data by size and position, and should also provide a high degree of regularity in order to support comparability by the analyst. In this paper, we present a framework for visualizing large sets of time series. Based on the notion of inter time series importance relationships, we define a set of objective functions that space-filling layout schemes for time series data should obey. We develop an efficient algorithm addressing the identified problems by generating layouts that reflect hierarchy and importance based relationships in a regular layout with favorable aspect ratios. We apply our technique to a number of real world data sets including sales and stock data, and we compare our technique with an aspect ratio aware variant of the well known TreeMap algorithm. The examples show the advantages and practical usefulness of our layout algorithm.

We present a system that allows users to interactively explore complex flow scenarios represented as Sankey diagrams. Our system provides an overview of the flow graph and allows users to zoom in and explore details on demand. The support for quantitative flow tracing across the flow graph as well as representations at different levels of detail facilitate the understanding of complex flow situations. The energy flow in a city serves as a sample scenario for our system. Different forms of energy are distributed within the city and they are transformed into heat, electricity, or other forms of energy. These processes are visualized and interactively explored. In addition our system can be used as a planning tool for the exploration of alternative scenarios by interactively manipulating different parameters in the energy flow network.

The general problem of visualizing "family trees", or genealogical graphs, in 2D, is considered. A graph theoretic analysis is given, which identifies why genealogical graphs can be difficult to draw. This motivates some novel graphical representations, including one based on a dual tree, a subgraph formed by the union of two trees. Dual trees can be drawn in various styles, including an indented outline style, and allow users to browse general multitrees in addition to genealogical graphs, by transitioning between different dual tree views. A software prototype for such browsing is described, that supports smoothly animated transitions, automatic camera framing, rotation of subtrees, and a novel interaction technique for expanding or collapsing subtrees to any depth with a single mouse drag

Existing system level taxonomies of visualization tasks are geared more towards the design of particular representations than the facilitation of user analytic activity. We present a set of ten low level analysis tasks that largely capture people's activities while employing information visualization tools for understanding data. To help develop these tasks, we collected nearly 200 sample questions from students about how they would analyze five particular data sets from different domains. The questions, while not being totally comprehensive, illustrated the sheer variety of analytic questions typically posed by users when employing information visualization systems. We hope that the presented set of tasks is useful for information visualization system designers as a kind of common substrate to discuss the relative analytic capabilities of the systems. Further, the tasks may provide a form of checklist for system designers.

Aggregating items can simplify the display of huge quantities of data values at the cost of losing information about the attribute values of the individual items. We propose a distribution glyph, in both two- and three-dimensional forms, which specifically addresses the concept of how the aggregated data is distributed over the possible range of values. It is capable of displaying distribution, variability and extent information for up to four attributes at a time of multivariate, clustered data. User studies validate the concept, showing that both glyphs are just as good as raw data and the 3D glyph is better for answering some questions.

The discrete nature of categorical data makes it a particular challenge for visualization. Methods that work very well for continuous data are often hardly usable with categorical dimensions. Only few methods deal properly with such data, mostly because of the discrete nature of categorical data, which does not translate well into the continuous domains of space and color. Parallel sets is a new visualization method that adopts the layout of parallel coordinates, but substitutes the individual data points by a frequency based representation. This abstracted view, combined with a set of carefully designed interactions, supports visual data analysis of large and complex data sets. The technique allows efficient work with meta data, which is particularly important when dealing with categorical datasets. By creating new dimensions from existing ones, for example, the user can filter the data according to his or her current needs. We also present the results from an interactive analysis of CRM data using parallel sets. We demonstrate how the flexible layout eases the process of knowledge crystallization, especially when combined with a sophisticated interaction scheme.

We present PRISAD, the first generic rendering infrastructure for information visualization applications that use the accordion drawing technique: rubber sheet navigation with guaranteed visibility for marked areas of interest. Our new rendering algorithms are based on the partitioning of screen space, which allows us to handle dense dataset regions correctly. The algorithms in previous work led to incorrect visual representations because of overculling, and to inefficiencies due to overdrawing multiple items in the same region. Our pixel based drawing infrastructure guarantees correctness by eliminating overculling, and improves rendering performance with tight bounds on overdrawing. PRITree and PRISeq are applications built on PRISAD, with the feature sets of TreeJuxtaposer and SequenceJuxtaposer, respectively. We describe our PRITree and PRISeq dataset traversal algorithms, which are used for efficient rendering, culling, and layout of datasets within the PRISAD framework. We also discuss PRITree node marking techniques, which offer order-of-magnitude improvements to both memory and time performance versus previous range storage and retrieval techniques. Our PRITree implementation features a five fold increase in rendering speed for nontrivial tree structures, and also reduces memory requirements in some real world datasets by up to eight times, so we are able to handle trees of several million nodes. PRISeq renders fifteen times faster and handles datasets twenty times larger than previous work.

In order to gain insight into multivariate data, complex structures must be analysed and understood. Parallel coordinates is an excellent tool for visualizing this type of data but has its limitations. This paper deals with one of its main limitations - how to visualize a large number of data items without hiding the inherent structure they constitute. We solve this problem by constructing clusters and using high precision textures to represent them. We also use transfer functions that operate on the high precision textures in order to highlight different aspects of the cluster characteristics. Providing predefined transfer functions as well as the support to draw customized transfer functions makes it possible to extract different aspects of the data. We also show how feature animation can be used as guidance when simultaneously analysing several clusters. This technique makes it possible to visually represent statistical information about clusters and thus guides the user, making the analysis process more efficient.

We present an effort to evaluate the possible utility of a new type of 3D glyphs intended for visualizations of multivariate spatial data. They are based on results from vision research suggesting that our perception of metric 3D structure is distorted and imprecise relative to the actual scene before us (e.g., "metric 3D structure in visualizations" by M. Lind et al. (2003)); only a class of qualitative properties of the scene is perceived with accuracy. These properties are best characterized as being invariant over affine but not Euclidean transformations. They are related, but not identical to, the non-accidental properties (NAPs) described by Lowe in "perceptual organization and visual recognition" (1984) on which the notion of geons is based in "recognition by components - a theory of image understanding" by I. Biederman (1987). A large number of possible 3D glyphs for the visualization of spatial data can be constructed using such properties. One group is based on the local sign of surface curvature. We investigated these properties in a visualization experiment. The results are promising and the implications for visualization are discussed.

Partitioning of geo-spatial data for efficient allocation of resources such as schools and emergency health care services is driven by a need to provide better and more effective services. Partitioning of spatial data is a complex process that depends on numerous factors such as population, costs incurred in deploying or utilizing resources and target capacity of a resource. Moreover, complex data such as population distributions are dynamic i.e. they may change over time. Simple animation may not effectively show temporal changes in spatial data. We propose the use of three temporal visualization techniques -wedges, rings and time slices - to display the nature of change in temporal data in a single view. Along with maximizing resource utilization and minimizing utilization costs, a partition should also ensure the long term effectiveness of the plan. We use multi-attribute visualization techniques to highlight the strengths and identify the weaknesses of a partition. Comparative visualization techniques allow multiple partitions to be viewed simultaneously. Users can make informed decisions about how to partition geo spatial data by using a combination of our techniques for multi-attribute visualization, temporal visualization and comparative visualization.

We present the Visual Code Navigator, a set of three interrelated visual tools that we developed for exploring large source code software projects from three different perspectives, or views: the syntactic view shows the syntactic constructs in the source code. The symbol view shows the objects a file makes available after compilation, such as function signatures, variables, and namespaces. The evolution view looks at different versions in a project lifetime of a number of selected source files. The views share one code model, which combines hierarchical syntax based and line based information from multiple source files versions. We render this code model using a visual model that extends the pixel-filling, space partitioning properties of shaded cushion treemaps with novel techniques. We discuss how our views allow users to interactively answer complex questions on various code elements by simple mouse clicks. We validate the efficiency and effectiveness of our toolset by an informal user study on the source code of VTK, a large, industry-size C++ code base

As information visualization matures as an academic research field, commercial spinoffs are proliferating, but success stories are harder to find. This is the normal process of emergence for new technologies, but the panel organizers believe that there are certain strategies that facilitate success. To teach these lessons, we have invited several key figures who are seeking to commercialize information visualization tools. The panelists make short presentations, engage in a moderated discussion, and respond to audience questions.

Many visual analysis tools operate on a fixed set of data. However, professional information analysts follow issues over a period of time and need to be able to easily add new documents to an ongoing exploration. Some analysts handle documents in a moving window of time, with new documents constantly added and old ones aging out. This paper describes both the user interaction and the technical implementation approach for a visual analysis system designed to support constantly evolving text collections.

A new pseudo coloring technique for large scale one-dimensional datasets is proposed. For visualization of a large scale dataset, user interaction is indispensable for selecting focus areas in the dataset. However, excessive switching of the visualized image makes it difficult for the user to recognize overview/ detail and detail/ detail relationships. The goal of this research is to develop techniques for visualizing details as precisely as possible in overview display. In this paper, visualization of a one-dimensional but very large dataset is considered. The proposed method is based on pseudo coloring, however, each scalar value corresponds to two discrete colors. By painting with two colors at each value, users can read out the value precisely. This method has many advantages: it requires little image space for visualization; both the overview and details of the dataset are visible in one image without distortion; and implementation is very simple. Several application examples, such as meteorological observation data and train convenience evaluation data, show the effectiveness of the method.

We present a novel visual correlation paradigm for situational awareness (SA) and suggest its usage in a diverse set of applications that require a high level of SA. Our approach is based on a concise and scalable representation, which leads to a flexible visualization tool that is both clear and intuitive to use. Situational awareness is the continuous extraction of environmental information, its integration with previous knowledge to form a coherent mental picture, and the use of that picture in anticipating future events. In this paper we build on our previous work on visualization for network intrusion detection and show how that approach can be generalized to encompass a much broader class of SA systems. We first propose a generalization that is based on what we term, the w3 premise, namely that each event must have at least the what, when and where attributes. We also present a second generalization, which increases flexibility and facilitates complex visual correlations. Finally, we demonstrate the generality of our approaches by applying our visualization paradigm in a collection of diverse SA areas.

The most common approach to support analysis of graphs with associated time series data include: overlay of data on graph vertices for one timepoint at a time by manipulating a visual property (e.g. color) of the vertex, along with sliders or some such mechanism to animate the graph for other timepoints. Alternatively, data from all the timepoints can be overlaid simultaneously by embedding small charts into graph vertices. These graph visualizations may also be linked to other visualizations (e.g., parallel co-ordinates) using brushing and linking. This paper describes a study performed to evaluate and rank graph+timeseries visualization options based on users' performance time and accuracy of responses on predefined tasks. The results suggest that overlaying data on graph vertices one timepoint at a time may lead to more accurate performance for tasks involving analysis of a graph at a single timepoint, and comparisons between graph vertices for two distinct timepoints. Overlaying data simultaneously for all the timepoints on graph vertices may lead to more accurate and faster performance for tasks involving searching for outlier vertices displaying different behavior than the rest of the graph vertices for all timepoints. Single views have advantage over multiple views on tasks that require topological information. Also, the number of attributes displayed on nodes has a non trivial influence on accuracy of responses, whereas the number of visualizations affect the performance time.

Exploratory visualization environments allow users to build and browse coordinated multiview visualizations interactively. As the number of views and amount of coordination increases, conceptualizing coordination structure becomes more and more important for successful data exploration. Integrated metavisualization is exploratory visualization of coordination and other interactive structure directly inside a visualization's own user interface. This paper presents a model of integrated metavisualization, describes the problem of capturing dynamic interface structure as visualizable data, and outlines three general approaches to integration. Metavisualization has been implemented in improvise, using views, lenses, and embedding to reveal the dynamic structure of its own highly coordinated visualizations.

Recent years have witnessed the dramatic popularity of online social networking services, in which millions of members publicly articulate mutual "friendship" relations. Guided by ethnographic research of these online communities, we have designed and implemented a visualization system for playful end-user exploration and navigation of large scale online social networks. Our design builds upon familiar node link network layouts to contribute customized techniques for exploring connectivity in large graph structures, supporting visual search and analysis, and automatically identifying and visualizing community structures. Both public installation and controlled studies of the system provide evidence of the system's usability, capacity for facilitating discovery, and potential for fun and engaged social activity

Treemaps are a well known method for the visualization of attributed hierarchical data. Previously proposed treemap layout algorithms are limited to rectangular shapes, which cause problems with the aspect ratio of the rectangles as well as with identifying the visualized hierarchical structure. The approach of Voronoi treemaps presented in this paper eliminates these problems through enabling subdivisions of and in polygons. Additionally, this allows for creating treemap visualizations within areas of arbitrary shape, such as triangles and circles, thereby enabling a more flexible adaptation of treemaps for a wider range of applications.

Analysis of degenerate tensors is a fundamental step in finding the topological structures and separatrices in tensor fields. Previous work in this area have been limited to analyzing symmetric second order tensor fields. In this paper, we extend the topological analysis to 2D general (asymmetric) second order tensor fields. We show that it is not sufficient to define degeneracies based on eigenvalues alone, but one must also include the eigenvectors in the analysis. We also study the behavior of these eigenvectors as they cross from one topological region into another.