IEEE VIS Publication Dataset

Adaptive search systems apply user models to provide better separation of relevant and non-relevant documents in a list of results. This paper presents our attempt to leverage this ability of user models in the context of visual information analysis. We developed an adaptive visualization approach for presentation and exploration of search results. We simulated a visual intelligence search/analysis scenario with log data extracted from an adaptive information foraging study and were able to verify that our method can improve the ability of traditional relevance visualization to separate relevant and irrelevant information.

The Geospatial Visual Analytics Toolkit intended for exploratory analysis of spatial and spatio-temporal data has been recently enriched with specific visual and computational techniques supporting analysis of data about movement. We applied these and other techniques to the data and tasks of Mini Challenge 4, where it was necessary to analyze tracks of moving people.CR Categories and Subject Descriptors: H.1.2 [User/Machine Systems]: Human information processing - Visual Analytics; 1.6.9 [Visualization]: information visualization.

Staining is a technique for categorizing time-varying spatial data; that is, data of things moving through space over time. In Staining, a stain is applied in either time or space, and the objects which move through the stain become marked. This technique and a research prototype demonstrating the technique were developed in response to the VAST 2008 Contest Mini-challenge: Evacuation Traces.

The adoption of visual analytics methodologies in security applications is an approach that could lead to interesting results. Usually, the data that has to be analyzed finds in a graphical representation its preferred nature, such as spatial or temporal relationships. Due to the nature of these applications, it is very important that key-details are made easy to identify. In the context of the VAST 2008 Challenge, we developed a visualization tool that graphically displays the movement of 82 employees of the Miami Department of Health (USA). We also asked 13 users to identify potential suspects and observe what happened during an evacuation of the building caused by an explosion. In this paper we explain the results of the user testing we conducted and how the users interpreted the event taken into account.

It has been widely accepted that interactive visualization techniques enable users to more effectively form hypotheses and identify areas for more detailed investigation. There have been numerous empirical user studies testing the effectiveness of specific visual analytical tools. However, there has been limited effort in connecting a userpsilas interaction with his reasoning for the purpose of extracting the relationship between the two. In this paper, we present an approach for capturing and analyzing user interactions in a financial visual analytical tool and describe an exploratory user study that examines these interaction strategies. To achieve this goal, we created two visual tools to analyze raw interaction data captured during the user session. The results of this study demonstrate one possible strategy for understanding the relationship between interaction and reasoning both operationally and strategically.

We seek an information-revealing representation for high-dimensional data distributions that may contain local trends in certain subspaces. Examples are data that have continuous support in simple shapes with identifiable branches. Such data can be represented by a graph that consists of segments of locally fit principal curves or surfaces summarizing each identifiable branch. We describe a new algorithm to find the optimal paths through such a principal graph. The paths are optimal in the sense that they represent the longest smooth trends through the data set, and jointly they cover the data set entirely with minimum overlap. The algorithm is suitable for hypothesizing trends in high-dimensional data, and can assist exploratory data analysis and visualization.

GeoTime and nSpace2 are interactive visual analytics tools that were used to examine and interpret all four of the 2008 VAST Challenge datasets. GeoTime excels in visualizing event patterns in time and space, or in time and any abstract landscape, while nSpace2 is a web-based analytical tool designed to support every step of the analytical process. nSpace2 is an integrating analytic environment. This paper highlights the VAST analytical experience with these tools that contributed to the success of these tools and this team for the third consecutive year.

The VAST 2008 Challenge consisted of four heterogeneous synthetic data sets each organized into separate mini-challenges. The Grand Challenge required integrating the raw data from these four data sets as well as integrating results and findings from team members working on specific mini-challenges. Modeling the problem with a semantic network provided a means for integrating both the raw data and the subjective findings.

Palantir is a world-class analytic platform used worldwide by governmental and financial analysts. This paper provides an introduction to the platform contextualized by its application to the 2008 IEEE VAST contest. In this challenge, we explored a notional dataset about a fabricated religious movement, Catalanopsilas Paraiso Manifesto Movement.

From instant messaging and email to wikis and blogs, millions of individuals are generating content that reflects their relationships with others in the world, both online and offline. Since communication artifacts are recordings of life events, we can gain insights into the social attributes and structures of the people within this communication history. In this paper, we describe SocialRank, an ego- and time-centric workflow for identifying social relationships in an email corpus. This workflow includes four high-level tasks: discovery, validation, annotation and dissemination. SocialRank combines relationship ranking algorithms with timeline, social network diagram, and multidimensional scaling visualization techniques to support these tasks.

Climate change has emerged as one of the grand global challenges facing humanity. The dominant anthropogenic greenhouse gas that seems to be contributing to the climate change problem, carbon dioxide (CO2), has a complex cycle through the atmosphere, oceans and biosphere. The combustion of fossil fuels (power production, transportation, etc.) remains the largest source of anthropogenic CO2 to the Earthpsilas atmosphere. Up until very recently, the quantification of fossil fuel CO2 was understood only at coarse space and time scales. A recent research effort has greatly improved this space/time quantification resulting in source data at a resolution of less than 10 km2/hr at the surface of North America. By providing visual tools to examine this new, high resolution CO2 data, we can better understand the way that CO2 is transmitted within the atmosphere and how it is exchanged with other components of the Earth System. We have developed interactive visual analytic tools, which allows for easy data manipulation, analysis, and extraction. The visualization system is aimed for a wide range of users which include researchers and political leaders. The goal is to help assist these people in analyzing data and enabling new policy options in mitigation of fossil fuel CO2 emissions in the U.S.

We present a process for the exploration and analysis of large databases of events. A typical database is characterized by the sequential actions of a number of individual entities. These entities can be compared by their similarities in sequence and changes in sequence over time. The correlation of two sequences can provide important clues as to the possibility of a connection between the responsible entities, but an analyst might not be able to specify the type of connection sought prior to examination. Our process incorporates extensive automated calculation and data mining but permits diversity of analysis by providing visualization of results at multiple levels, taking advantage of human intuition and visual processing to generate avenues of inquiry.

The speed of data retrieval qualitatively affects how analysts visually explore and analyze their data. To ensure smooth interactions in massive time series datasets, one needs to address the challenges of computing ad hoc queries, distributing query load, and hiding system latency. In this paper, we present ATLAS, a visualization tool for temporal data that addresses these issues using a combination of high performance database technology, predictive caching, and level of detail management. We demonstrate ATLAS using commodity hardware on a network traffic dataset of more than a billion records.

The SPADAC team used various visual analytics tools and methods to find geo-temporal patterns of migration from a Caribbean island from 2005-2007. In this paper, we describe the tools and methods used in the analysis. These methods included generating temporal variograms, dendrograms, and proportionally weighted migration maps, using tools such as the R statistical software package and Signature Analysttrade. We found that there is a significant positive space-time correlation with the boat encounters (especially the landings), with a migratory shift further away from the point of departure over time.

We provide a description of the tools and techniques used in our analysis of the VAST 2008 Challenge dealing with mass movement of persons departing Isla Del Sue.no on boats for the United States during 2005-2007. We used visual analytics to explore migration patterns, characterize the choice and evolution of landing sites, characterize the geographical patterns of interdictions and determine the successful landing rate. Our ComVis tool, in connection with some helper applications and Google Earth, allowed us to explore geo-temporal characteristics of the data set and answer the challenge questions. The ComVis project file captures the visual analysis context and facilitates better collaboration among team members.

We describe a visual analytics (VA) infrastructure, rooted on techniques in machine learning and logic-based deductive reasoning that will assist analysts to make sense of large, complex data sets by facilitating the generation and validation of models representing relationships in the data. We use logic programming (LP) as the underlying computing machinery to encode the relations as rules and facts and compute with them. A unique aspect of our approach is that the LP rules are automatically learned, using Inductive Logic Programming, from examples of data that the analyst deems interesting when viewing the data in the high-dimensional visualization interface. Using this system, analysts will be able to construct models of arbitrary relationships in the data, explore the data for scenarios that fit the model, refine the model if necessary, and query the model to automatically analyze incoming (future) data exhibiting the encoded relationships. In other words it will support both model-driven data exploration, as well as data-driven model evolution. More importantly, by basing the construction of models on techniques from machine learning and logic-based deduction, the VA process will be both flexible in terms of modeling arbitrary, user-driven relationships in the data as well as readily scale across different data domains.

Analysis of multidimensional data often requires careful examination of relationships across dimensions. Coordinated multiple view approaches have become commonplace in visual analysis tools because they directly support expression of complex multidimensional queries using simple interactions. However, generating such tools remains difficult because of the need to map domain-specific data structures and semantics into the idiosyncratic combinations of interdependent data and visual abstractions needed to reveal particular patterns and distributions in cross-dimensional relationships. This paper describes: (1) a method for interactively expressing sequences of multidimensional set queries by cross-filtering data values across pairs of views, and (2) design strategies for constructing coordinated multiple view interfaces for cross-filtered visual analysis of multidimensional data sets. Using examples of cross-filtered visualizations of data from several different domains, we describe how cross-filtering can be modularized and reused across designs, flexibly customized with respect to data types across multiple dimensions, and incorporated into more wide-ranging multiple view designs. The demonstrated analytic utility of these examples suggest that cross-filtering is a suitable design pattern for instantiation in a wide variety of visual analysis tools.

Understanding multivariate relationships is an important task in multivariate data analysis. Unfortunately, existing multivariate visualization systems lose effectiveness when analyzing relationships among variables that span more than a few dimensions. We present a novel multivariate visual explanation approach that helps users interactively discover multivariate relationships among a large number of dimensions by integrating automatic numerical differentiation techniques and multidimensional visualization techniques. The result is an efficient workflow for multivariate analysis model construction, interactive dimension reduction, and multivariate knowledge discovery leveraging both automatic multivariate analysis and interactive multivariate data visual exploration. Case studies and a formal user study with a real dataset illustrate the effectiveness of this approach.

Analyzing unstructured text streams can be challenging. One popular approach is to isolate specific themes in the text, and to visualize the connections between them. Some existing systems, like ThemeRiver, provide a temporal view of changes in themes; other systems, like In-Spire, use clustering techniques to help an analyst identify the themes at a single point in time. Narratives combines both of these techniques; it uses a temporal axis to visualize ways that concepts have changed over time, and introduces several methods to explore how those concepts relate to each other. Narratives is designed to help the user place news stories in their historical and social context by understanding how the major topics associated with them have changed over time. Users can relate articles through time by examining the topical keywords that summarize a specific news event. By tracking the attention to a news article in the form of references in social media (such as weblogs), a user discovers both important events and measures the social relevance of these stories.

Data about movements of various objects are collected in growing amounts by means of current tracking technologies. Traditional approaches to visualization and interactive exploration of movement data cannot cope with data of such sizes. In this research paper we investigate the ways of using aggregation for visual analysis of movement data. We define aggregation methods suitable for movement data and find visualization and interaction techniques to represent results of aggregations and enable comprehensive exploration of the data. We consider two possible views of movement, traffic-oriented and trajectory-oriented. Each view requires different methods of analysis and of data aggregation. We illustrate our argument with example data resulting from tracking multiple cars in Milan and example analysis tasks from the domain of city traffic management.