This PDF file contains the front matter associated with SPIE Proceedings Volume 7530, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

Fuzzy sets refer to the datasets which do not have separate, distinct clusters, and they contain data elements whose membership values are between 0 and 1. Often each element belongs to several groups. We design a disk diagram with an augmented frequency graph to visualize elements in a fuzzy set. It shows the distribution of the elements within the set with respect to the membership function value. Fuzzy set operations-like intersection-are used to find the elements of interests with certain degree of confidence. However, few techniques exist for visualization of fuzzy set operations whose process and result provide both the relationship and the distribution of individual data point between sets. With the proposed disk diagram representing a fuzzy set, we suggest an interactive visualization system to analyze fuzzy data and grasp the relationship among the sets. By interactively overlapping the disk diagrams, a user can not only bring out the individual elements of interests, but also overview the whole data elements associated with the sets. Furthermore, we investigated two different visualization scenarios for intuitive interpretation of the visualization results. We demonstrate the proposed visualization system with a terrorist analysis output dataset.

Parallel coordinates technique has been widely used in information visualization applications and it has achieved great success in visualizing multivariate data and perceiving their trends. Nevertheless, visual clutter usually weakens or even diminishes its ability when the data size increases. In this paper, we first propose a tile-based parallel coordinates, where the plotting area is divided into rectangular tiles. Each tile stores an intersection density that counts the total number of polylines intersecting with that tile. Consequently, the intersection density is mapped to optical attributes, such as color and opacity, by interactive transfer functions. The method visualizes the polylines efficiently and informatively in accordance with the density distribution, and thus, reduces visual cluttering and promotes knowledge discovery. The interactivity of our method allows the user to instantaneously manipulate the tiles distribution and the transfer functions. Specifically, the classic parallel coordinates rendering is a special case of our method when each tile represents only one pixel. A case study on a real world data set, U.S. stock mutual fund data of year 2006, is presented to show the capability of our method in visually analyzing financial data. The presented visual analysis is conducted by an expert in the domain of finance. Our method gains the support from professionals in the finance field, they embrace it as a potential investment analysis tool for mutual fund managers, financial planners, and investors.

We present a system for visualizing magnetic resonance spectroscopy (MRS) data sets. Using MRS, radiologists generate multiple 3D scalar fields of metabolite concentrations within the brain and compare them to anatomical magnetic resonance imaging. By understanding the relationship between metabolic makeup and anatomical structure, radiologists hope to better diagnose and treat tumors and lesions. Our system consists of three linked visualizations: a spatial glyph-based technique we call Scaled Data-Driven Spheres, a parallel coordinates visualization augmented to incorporate uncertainty in the data, and a slice plane for accurate data value extraction. The parallel coordinates visualization uses specialized brush interactions designed to help users identify nontrivial linear relationships between scalar fields. We describe two novel contributions to parallel coordinates visualizations: linear function brushing and new axis construction. Users have discovered significant relationships among metabolites and anatomy by linking interactions between the three visualizations.

The effect of harmonious versus non-harmonious color combinations in one aspect of information visualization effectiveness is considered. One focus is the relative suitability of two competing paradigms for determining color combinations that are harmonious. A second focus is the suitability of using opposing (i.e., opponent) colors for feature presentation in information visualization. The effects are considered for color item overlays on crowded and noncrowded displays. A statistical analysis of human responses is also presented.

The scatter plot is a well-known method of visualizing pairs of two-dimensional continuous variables. Multidimensional data can be depicted in a scatter plot matrix. They are intuitive and easy-to-use, but often have a high degree of overlap which may occlude a significant portion of data. In this paper, we propose variable binned scatter plots to allow the visualization of large amounts of data without overlapping. The basic idea is to use a non-uniform (variable) binning of the x and y dimensions and plots all the data points that fall within each bin into corresponding squares. Further, we map a third attribute to color for visualizing clusters. Analysts are able to interact with individual data points for record level information. We have applied these techniques to solve real-world problems on credit card fraud and data center energy consumption to visualize their data distribution and cause-effect among multiple attributes. A comparison of our methods with two recent well-known variants of scatter plots is included.

In this paper we introduce methods for the visualization of ontologies using different geometrical representations. An ontology is a formal way to define domain knowledge by means of axioms about domain concepts, properties and individuals. Currently, ontologies are modeled with the OWL language; this language is very expressive and provides challenges for ontology visualization. Expressive ontologies can be difficult to understand and to that end ontology visualization can be extremely helpful for ontology inspection during the process of development as well as for inspection of existing ontologies. Our improved approach for ontology visualization includes two different tree-visualization techniques: i.e., the node-link technique and the containment technique. The node-link technique visualizes the ontology as a graph. The graph can be build for each concept with different levels of depth. The core visualization component is based on the spanning tree skeleton of the graph and it includes five different geometrical views, i.e., two Euclidean, two hyperbolic and one spherical. All the views are augmented with corresponding geometrical transformations so that user interaction like pan, zoom and rotate can be invoked. Another approach encompasses a 3-dimensional spherical alternative of the treemap method, in which nodes are placed on the surface of a sphere. Each parental node contains its children, which are places on the surface of the parent. We augmented this method with semantic zoom technique. With this technique the level of details depends on the distance from the viewer. Our approach provides the means to visualize ontology from different perspectives and different levels of detail. The interaction that is provided greatly enhances the user perception of otherwise complex information.

The exploration of multidimensional scalar fields is commonly based on the knowledge of the topology of their isosurfaces. The latter is established through the analysis of critical regions of the studied fields. A new method, based on homology theory, for the detection and classification of critical regions in multidimensional scalar fields is proposed in this paper. The use of computational homology provides an efficient and successful algorithm that works in all dimensions and allows to generalize visual classification techniques based solely on the notion of connectedness which appears insufficient in higher dimensions. We present the algorithm, discuss details of its implementation, and illustrate it by experimentations in two, three, and four dimensional spaces.

Iterative clustering (e.g. K-Means, EM) is one of the most commonly used clustering methods, which attempts to iteratively find a local optimum starting from an initial condition, including initial centroids and initial number of clusters. For iterative clustering, research has shown that the initial conditions are crucial to clustering quality and running time of a clustering computation. Using a novel visualization tool, CComViz (Cluster Comparison Visualization), we present an innovative approach to refine the initial centroids and the number of clusters by visually analyzing multiple clustering results generated by different clustering algorithms. As an example, we apply our new approach to a gene expression case study for generating a better and converging clustering. The proposed approach is considered to be an extension to cluster ensembles since the original data sources are reused, while in classic cluster ensembles they are not.

The design of transfer functions for volume rendering is a difficult task. This is particularly true for multichannel data sets, where multiple data values exist for each voxel. In this paper, we propose a new method for transfer function design. Our new method provides a framework to combine multiple approaches and pushes the boundary of gradient-based transfer functions to multiple channels, while still keeping the dimensionality of transfer functions to a manageable level, i.e., a maximum of three dimensions, which can be displayed visually in a straightforward way. Our approach utilizes channel intensity, gradient, curvature and texture properties of each voxel. The high-dimensional data of the domain is reduced by applying recently developed nonlinear dimensionality reduction algorithms. In this paper, we used Isomap as well as a traditional algorithm, Principle Component Analysis (PCA). Our results show that these dimensionality reduction algorithms significantly improve the transfer function design process without compromising visualization accuracy. In this publication we report on the impact of the dimensionality reduction algorithms on transfer function design for confocal microscopy data.

This paper presents the concept of Monotone Boolean Function Visual Analytics (MBFVA) and its application to the medical domain. The medical application is concerned with discovering breast cancer diagnostic rules (i) interactively with a radiologist, (ii) analytically with data mining algorithms, and (iii) visually. The coordinated visualization of these rules opens an opportunity to coordinate the rules, and to come up with rules that are meaningful for the expert in the field, and are confirmed with the database. This paper shows how to represent and visualize binary multivariate data in 2-D and 3-D. This representation preserves the structural relations that exist in multivariate data. It creates a new opportunity to guide the visual discovery of unknown patterns in the data. In particular, the structural representation allows us to convert a complex border between the patterns in multidimensional space into visual 2-D and 3-D forms. This decreases the information overload on the user. The visualization shows not only the border between classes, but also shows a location of the case of interest relative to the border between the patterns. A user does not need to see the thousands of previous cases that have been used to build a border between the patterns. If the abnormal case is deeply inside in the abnormal area, far away from the border between "normal" and "abnormal" patterns, then this shows that this case is very abnormal and needs immediate attention. The paper concludes with the outline of the scaling of the algorithm for the large data sets.

The analysis of binary data remains a challenge, especially for large or potentially inconsistent files. Traditionally, hex editors only make limited use of semantic information available to the user. We present an editor that supports user-supplied semantic data definitions. This semantic information is used throughout the program to realize semantic data visualization and data exploration capabilities not present in similar systems. Visualization and human-computer interaction techniques are applied. We show that this makes recognizing the structure of unknown or inconsistent data much more effective. Our approach demonstrates concepts that can be applied to the visual analysis of raw data in general.

The analysis of high-dimensional data is an important, yet inherently difficult problem. Projection techniques such as PCA, MDS, and SOM can be used to map high-dimensional data to 2D display space. However, projections typically incur a loss in information. Often uncertainty exists regarding the precision of the projection as compared with its original data characteristics. While the output quality of these projection techniques can be discussed in terms of algorithmic assessment, visualization is often helpful for better understanding the results. We address the visual assessment of projection precision by an approach integrating an appropriately designed projection precision measure directly into the projection visualization. To this end, a flexible projection precision measure is defined that allows the user to balance the degree of locality at which the measure is evaluated. Several visual mappings are designed for integrating the precision measure into the projection visualization at various levels of abstraction. The techniques are implemented in a fully interactive system which is practically applied on several data sets. We demonstrate the usefulness of the approach for visual analysis of classified and clustered high-dimensional data sets. We thereby show how our novel interactive precision quality visualization system helps to examine preservation of closeness of the data in original space into the low-dimensional space.

While there have been intensive efforts in developing better 3D flow visualization techniques, little attention has been paid to the design of better user interfaces and more effective data exploration work flow. In this paper, we propose a novel graph-based user interface called Flow Web to enable more systematic explorations of 3D flow data. The Flow Web is a node-link graph that is constructed to highlight the essential flow structures where a node represents a region in the field and a link connects two nodes if there exist particles traveling between the regions. The direction of an edge implies the flow path, and the weight of an edge indicates the number of particles traveling through the connected nodes. Hierarchical flow webs are created by splitting or merging nodes and edges to allow for easy understanding of the underlying flow structures. To draw the Flow Web, we adopt force based graph drawing algorithms to minimize edge crossings, and use a hierarchical layout to facilitate the study of flow patterns step by step. The Flow Web also supports user queries to the properties of nodes and links. Examples of the queries for node properties include the degrees, complexity, and some associated physical attributes such as velocity magnitude. Queries for edges include weights, flow path lengths, existence of circles and so on. It is also possible to combine multiple queries using operators such as and , or, not. The FlowWeb supports several types of user interactions. For instance, the user can select nodes from the subgraph returned by a query and inspect the nodes with more details at different levels of detail. There are multiple advantages of using the graph-based user interface. One is that the user can identify regions of interest much more easily since, unlike inspecting 3D regions, there is very little occlusion. It is also much more convenient for the user to query statistical information about the nodes and links at different levels of detail. With the Flow Web, it becomes easier for the user to log and track the progress of data exploration which is crucial for exploring large data sets. We demonstrate how to construct and draw the Flow Web effectively, and how to query the Flow Web to retrieve useful information from the data. Case studies are provided to demonstrate the exploration process.

Commercial websites offer many items to potential site users. However, most current websites display results of a search in text lists, or as lists sorted on one or two single criteria. Finding the best item in a text list based on multi-priority criteria is an exhausting task, especially for long lists. Visualizing search results and enabling users to perceive the tradeoffs among the results based on multiple priorities may ease this process. To investigate this, two different techniques for displaying and sorting search results are studied in this paper; Text, and XY Iconic Visualization. The goal is to determine which technique for representing search results would be the most efficient one for a website user. We conducted a user study to compare the usability of the two techniques. Collected data is in the form of participants' task responses, a satisfaction questionnaire, qualitative observations, and participants' comments. According to the results, iconic visualization is better for overview (it gives a good overview in a short amount of time) and search with more than two criteria, while text-based performs better for displaying details.

The goal of our project is to create a set of next-generation cyber situational-awareness capabilities with applications to other domains in the long term. The situational-awareness capabilities being developed focus on novel visualization techniques as well as data analysis techniques designed to improve the comprehensibility of the visualizations. The objective is to improve the decision-making process to enable decision makers to choose better actions. To this end, we put extensive effort into ensuring we had feedback from network analysts and managers and understanding what their needs truly are. This paper discusses the cognitive task analysis methodology we followed to acquire feedback from the analysts. This paper also provides the details we acquired from the analysts on their processes, goals, concerns, etc. A final result we describe is the generation of a task-flow diagram.

This paper presents MetroMap, a new graphical representation model for controlling and managing the software development process. Metromap uses metaphors and visual representation techniques to explore several key indicators in order to support problem detection and resolution. The resulting visualization addresses diverse management tasks, such as tracking of deviations from the plan, analysis of patterns of failure detection and correction, overall assessment of change management policies, and estimation of product quality. The proposed visualization uses a metaphor with a metro map along with various interactive techniques to represent information concerning the software development process and to deal efficiently with multivariate visual queries. Finally, the paper shows the implementation of the tool in JavaFX with data of a real project and the results of testing the tool with the aforementioned data and users attempting several information retrieval tasks. The conclusion shows the results of analyzing user response time and efficiency using the MetroMap visualization system. The utility of the tool was positively evaluated.

Design patterns have proven to be a useful means to make the process of designing, developing, and reusing software systems more efficient. In the area of information visualization, researchers have proposed design patterns for different functional components of the visualization pipeline. Since many visualization techniques need to display derived data as well as raw data, the data transformation stage is very important in the pipeline, yet existing design patterns are, in general, not sufficient to implement these data transformation techniques. In this paper, we propose two design patterns, operatorcentric transformation and data modifier, to facilitate the design of data transformations for information visualization systems. The key idea is to use operators to describe the data derivation and introduce data modifiers to represent the derived data. We also show that many interaction techniques can be regarded as operators as defined here, thus these two design patterns could support a wide range of visualization techniques. In addition, we describe a third design pattern, modifier-based visual mapping, that can generate visual abstraction via linking data modifiers to visual attributes. We also present a framework based on these three design patterns that supports coordinated multiple views. Several examples of multivariate visualizations are discussed to show that our design patterns and framework can improve the reusability and extensibility of information visualization systems. Finally, we explain how we have ported an existing visualization tool (XmdvTool) from its old data-centric structure to a new structure based on the above design patterns and framework.

An architecture for distributed and collaborative visualization is presented. The design goals of the system are to create a lightweight, easy to use and extensible framework for reasearch in scientific visualization. The system provides both single user and collaborative distributed environment. System architecture employs a client-server model. Visualization projects can be synchronously accessed and modified from different client machines. We present a set of visualization use cases that illustrate the flexibility of our system. The framework provides a rich set of reusable components for creating new applications. These components make heavy use of leading design patterns. All components are based on the functionality of a small set of interfaces. This allows new components to be integrated seamlessly with little to no effort. All user input and higher-level control functionality interface with proxy objects supporting a concrete implementation of these interfaces. These light-weight objects can be easily streamed across the web and even integrated with smart clients running on a user's cell phone. The back-end is supported by concrete implementations wherever needed (for instance for rendering). A middle-tier manages any communication and synchronization with the proxy objects. In addition to the data components, we have developed several first-class GUI components for visualization. These include a layer compositor editor, a programmable shader editor, a material editor and various drawable editors. These GUI components interact strictly with the interfaces. Access to the various entities in the system is provided by an AssetManager. The asset manager keeps track of all of the registered proxies and responds to queries on the overall system. This allows all user components to be populated automatically. Hence if a new component is added that supports the IMaterial interface, any instances of this can be used in the various GUI components that work with this interface. One of the main features is an interactive shader designer. This allows rapid prototyping of new visualization renderings that are shader-based and greatly accelerates the development and debug cycle.

We present MTVis, a multi-touch interactive tree visualization system. The multi-touch interface display hardware is built using the LED-LP technology, and the tree layout is based on RINGS, but enhanced with multitouch interactions. We describe the features of the system, and how the multi-touch interface enhances the user's experience in exploring the tree data structure. In particular, the multi-touch interface allows the user to simultaneously control two child nodes of the root, and rotate them so that some nodes are magnified, while preserving the layout of the tree. We also describe the other meaninful touch screen gestures the users can use to intuitively explore the tree.

Spatialization is a special kind of visualization that projects multidimensional data into low-dimensional representational spaces by making use of spatial metaphors. Spatialization methods face a dual challenge: on the one hand, to apply dimension reduction techniques in order to overcome the limitations of the representational space, and on the other hand, to provide a metaphoric framework for the visualization of information at different levels of granularity. This paper investigates how granularity is modeled and visualized by the existing spatialization methods, and introduces a new approach based on kernel density estimation and landscape metaphor. According to our approach, clusters of multidimensional data are revealed by landscape "relief", and are hierarchically organized into different levels of granularity through landscape "smoothness." In addition, it is demonstrated, herein, how the exploration of information at different levels of granularity is supported by appropriate operations in the framework of an interactive spatialization environment prototype.

Turbulent flows play a critical role in many fields, yet our understanding of the fundamental physics of turbulence remains in its infancy. One of the long term goals in turbulence research is to develop an improved understanding of the dynamic evolution, interaction and organization of vortices in three-dimensional turbulent flow. However this task is complicated by the lack of a clear, mathematically precise definition of what a vortex is. We believe that the design of effective methods for vortex identification and segmentation in complicated turbulent flows can be facilitated by the clear, detailed visual presentation of the multiple scalar and vector quantities potentially relevant to the feature identification process. In this paper, we present several different methods aimed at facilitating the integrated understanding of a variety of local measures extracted from 3D multivariate flow data, including quantities, directions, and orientation. A key focus of our work is on the development of methods for illustrating the local relationships between scalar and vector values important to the vortex identification process such as vorticity, swirl, and velocity, along with their direction and magnitude. Our methods include the use of arrows and glyphs or 3D texture along with different color coding strategies. We demonstrate our methods on a range of data including 3D turbulent boundary flow data and time varying ring data. The variety of multi-variate visualization methods that we have developed has succeeded in supporting fluids researchers in their efforts to gain deeper insights into their data.

In this work, we describe our visualization approach for business processes using 2.5 dimensional techniques (2.5D). The idea of 2.5D is to add the concept of layering to a two dimensional (2D) visualization. The layers are arranged in a three-dimensional display space. For the modeling of the business processes, we use the Business Process Modeling Notation (BPMN). The benefit of connecting BPMN with a 2.5D visualization is not only to obtain a more abstract view on the business process models but also to develop layering criteria that eventually increase readability of the BPMN model compared to 2D. We present a 2.5D Navigator for BPMN models that offers different perspectives for visualization. Therefore we also develop BPMN specific perspectives. The 2.5D Navigator combines the 2.5D approach with perspectives and allows free navigation in the three dimensional display space. We also demonstrate our tool and libraries used for implementation of the visualizations. The underlying general framework for 2.5D visualizations is explored and presented in a fashion that it can easily be used for different applications. Finally, an evaluation of our navigation tool demonstrates that we can achieve satisfying and aesthetic displays of diagrams stating BPMN models in 2.5D-visualizations.