IEEE VIS Publication Dataset

Sources of streaming information, such as news syndicates, publish information continuously. Information portals and news aggregators list the latest information from around the world enabling information consumers to easily identify events in the past 24 hours. The volume and velocity of these streams causes information from prior days to quickly vanish despite its utility in providing an informative context for interpreting new information. Few capabilities exist to support an individual attempting to identify or understand trends and changes from streaming information over time. The burden of retaining prior information and integrating with the new is left to the skills, determination, and discipline of each individual. In this paper we present a visual analytics system for linking essential content from information streams over time into dynamic stories that develop and change over multiple days. We describe particular challenges to the analysis of streaming information and present a fundamental visual representation for showing story change and evolution over time.

The HRL anomaly analysis tool was developed as part of the IEEE VAST Challenge 2009. One of the tasks involved processing badge and network traffic in order to detect and identify a fictitious embassy employee suspected of leaking information. The tool is designed to assist an analyst in detecting, analyzing, and visualizing anomalies and their relationships. Two key visualizations in our submission present how we identified the suspicious traffic using network visualization and how subsequently we connected that activity to an employee using an alibi table.

In this article, I describe the tools and techniques used to generate competing hypotheses for the VAST 2009 Flitter mini challenge. I will describe how I approached solving the social networks and the importance of the geospatial relationships to determine that ldquoSocial Structure Form Ardquo was the best matching social network.

Despite the growing number of systems providing visual analytic support for investigative analysis, few empirical studies of the potential benefits of such systems have been conducted, particularly controlled, comparative evaluations. Determining how such systems foster insight and sensemaking is important for their continued growth and study, however. Furthermore, studies that identify how people use such systems and why they benefit (or not) can help inform the design of new systems in this area. We conducted an evaluation of the visual analytics system Jigsaw employed in a small investigative sensemaking exercise, and we compared its use to three other more traditional methods of analysis. Sixteen participants performed a simulated intelligence analysis task under one of the four conditions. Experimental results suggest that Jigsaw assisted participants to analyze the data and identify an embedded threat. We describe different analysis strategies used by study participants and how computational support (or the lack thereof) influenced the strategies. We then illustrate several characteristics of the sensemaking process identified in the study and provide design implications for investigative analysis tools based thereon. We conclude with recommendations for metrics and techniques for evaluating other visual analytics investigative analysis tools.

An increasing number of temporal categorical databases are being collected: Electronic Health Records in healthcare organizations, traffic incident logs in transportation systems, or student records in universities. Finding similar records within these large databases requires effective similarity measures that capture the searcher's intent. Many similarity measures exist for numerical time series, but temporal categorical records are different. We propose a temporal categorical similarity measure, the M&M (Match & Mismatch) measure, which is based on the concept of aligning records by sentinel events, then matching events between the target and the compared records. The M&M measure combines the time differences between pairs of events and the number of mismatches. To accom-modate customization of parameters in the M&M measure and results interpretation, we implemented Similan, an interactive search and visualization tool for temporal categorical records. A usability study with 8 participants demonstrated that Similan was easy to learn and enabled them to find similar records, but users had difficulty understanding the M&M measure. The usability study feedback, led to an improved version with a continuous timeline, which was tested in a pilot study with 5 participants.

FinVis is a visual analytics tool that allows the non-expert casual user to interpret the return, risk and correlation aspects of financial data and make personal finance decisions. This interactive exploratory tool helps the casual decision-maker quickly choose between various financial portfolio options and view possible outcomes. FinVis allows for exploration of inter-temporal data to analyze outcomes of short-term or long-term investment decisions. FinVis helps the user overcome cognitive limitations and understand the impact of correlation between financial instruments in order to reap the benefits of portfolio diversification. Because this software is accessible by non-expert users, decision-makers from the general population can benefit greatly from using FinVis in practical applications. We quantify the value of FinVis using experimental economics methods and find that subjects using the FinVis software make better financial portfolio decisions as compared to subjects using a tabular version with the same information. We also find that FinVis engages the user, which results in greater exploration of the dataset and increased learning as compared to a tabular display. Further, participants using FinVis reported increased confidence in financial decision-making and noted that they were likely to use this tool in practical application.

Cellular radio networks are continually growing in both node count and complexity. It therefore becomes more difficult to manage the networks and necessary to use time and cost effective automatic algorithms to organize the networks neighbor cell relations. There have been a number of attempts to develop such automatic algorithms. Network operators, however, may not trust them because they need to have an understanding of their behavior and of their reliability and performance, which is not easily perceived. This paper presents a novel Web-enabled geovisual analytics approach to exploration and understanding of self-organizing network data related to cells and neighbor cell relations. A demonstrator and case study are presented in this paper, developed in close collaboration with the Swedish telecom company Ericsson and based on large multivariate, time-varying and geospatial data provided by the company. It allows the operators to follow, interact with and analyze the evolution of a self-organizing network and enhance their understanding of how an automatic algorithm configures locally-unique physical cell identities and organizes neighbor cell relations of the network. The geovisual analytics tool is tested with a self-organizing network that is operated by the automatic neighbor relations (ANR) algorithm. The demonstrator has been tested with positive results by a group of domain experts from Ericsson and will be tested in production.

This paper demonstrates the promise of augmenting interactive multivariate representations with information from statistical processes in the domain of weather data analysis. Statistical regression, correlation analysis, and descriptive statistical calculations are integrated via graphical indicators into an enhanced parallel coordinates system, called the Multidimensional Data eXplorer (MDX). These statistical indicators, which highlight significant associations in the data, are complemented with interactive visual analysis capabilities. The resulting system allows a smooth, interactive, and highly visual workflow. The system's utility is demonstrated with an extensive hurricane climate study that was conducted by a hurricane expert. In the study, the expert used a new data set of environmental weather data, composed of 28 independent variables, to predict annual hurricane activity. MDX shows the Atlantic Meridional Mode increases the explained variance of hurricane seasonal activity by 7-15% and removes less significant variables used in earlier studies. The findings and feedback from the expert (1) validate the utility of the data set for hurricane prediction, and (2) indicate that the integration of statistical processes with interactive parallel coordinates, as implemented in MDX, addresses both deficiencies in traditional weather data analysis and exhibits some of the expected benefits of visual data analysis.

The VAST 2009 Challenge consisted of three heterogeneous synthetic data sets organized into separate mini-challenges with minimal correspondence information. The challenge task was the identification of a suspected data theft from cyber and real-world traces. The grand challenge required integrating the findings from the mini challenges into a plausible, consistent scenario. A mixture of linked, customized tools based on queryable models and rapid prototyping as well as generic analysis tools (developed in-house) helped us correctly solve all of the mini challenges. A collaborative analytic process was employed to reconstruct the scenario and to propose the correct steps for the reliable identification of the criminal organization based on activity traces of its members.

In the VAST Challenge 2009 suspicious behavior had to be detected applying visual analytics to heterogeneous data, such as network traffic, social network enriched with geo-spatial attributes, and finally video surveillance data. This paper describes some of the awarded parts from our solution entry.

Although many in the community have advocated user-centered evaluations for visual analytic environments, a significant barrier exists. The users targeted by the visual analytics community (law enforcement personnel, professional information analysts, financial analysts, health care analysts, etc.) are often inaccessible to researchers. These analysts are extremely busy and their work environments and data are often classified or at least confidential. Furthermore, their tasks often last weeks or even months. It is simply not feasible to do such long-term observations to understand their jobs. How then can we hope to gather enough information about the diverse user populations to understand their needs? Some researchers, including the author, have been successful in getting access to specific end-users. A reasonable approach, therefore, would be to find a way to share user information. This work outlines a proposal for developing a handbook of user profiles for use by researchers, developers, and evaluators.

The poster introduces interactive multiobjective optimization (IMO) as a field offering new application possibilities and challenges for visual analytics (VA), and aims at inspiring collaboration between the two fields. Our aim is to collect new ideas in order to be able to utilize VA techniques more effectively in our user interface development. Simulation-based IMO methods are developed for complex problem solving, where the expert decision maker (analyst) should be supported during the iterative process of eliciting preference information and examining the resulting output data. IMO is a subfield of multiple criteria decision making (MCDM). In simulation-based IMO, the optimization task is formulated in a mathematical model containing several conflicting objectives and constraints depending on decision variables. While using IMO methods the analyst progressively provides preference information in order to find the most satisfactory compromise between the conflicting objectives. In the poster, the implementations of two new IMO methods are used as examples to demonstrate concrete challenges of interaction design. One of them is described in this summary.

Interactive visualization methods are often used to aid in the analysis of large datasets. We present a novel interactive visualization technique designed specifically for the analysis of location reporting patterns within large time-series datasets. We use a set of triangles with color coding to indicate the time between location reports. This allows reporting patterns (expected and unexpected) to be easily discerned during interactive analysis. We discuss the details of our method and describe evaluation both from expert opinion and from a user study.

One of the most common operations in exploration and analysis of various kinds of data is clustering, i.e. discovery and interpretation of groups of objects having similar properties and/or behaviors. In clustering, objects are often treated as points in multi-dimensional space of properties. However, structurally complex objects, such as trajectories of moving entities and other kinds of spatio-temporal data, cannot be adequately represented in this manner. Such data require sophisticated and computationally intensive clustering algorithms, which are very hard to scale effectively to large datasets not fitting in the computer main memory. We propose an approach to extracting meaningful clusters from large databases by combining clustering and classification, which are driven by a human analyst through an interactive visual interface.

Patents are an important economic factor in todays globalized markets. Therefore, the analysis of patent information has become an inevitable task for a variety of interest groups. The retrieval of relevant patent information is an integral part of almost every patent analysis scenario. Unfortunately, the complexity of patent material inhibits a straightforward retrieval of all relevant patent documents and leads to iterative, time-consuming approaches in practice. With `PatViz', a new system for interactive analysis of patent information has been developed to leverage iterative query refinement. PatViz supports users in building complex queries visually and in exploring patent result sets interactively. Thereby, the visual query module introduces an abstraction layer that provides uniform access to different retrieval systems and relieves users of the burden to learn different complex query languages. By establishing an integrated environment it allows for interactive reintegration of insights gained from visual result set exploration into the visual query representation. We expect that the approach we have taken is also suitable to improve iterative query refinement in other Visual Analytics systems.

Latent Semantic Analysis (LSA) is a commonly-used method for automated processing, modeling, and analysis of unstructured text data. One of the biggest challenges in using LSA is determining the appropriate model parameters to use for different data domains and types of analyses. Although automated methods have been developed to make rank and scaling parameter choices, these approaches often make choices with respect to noise in the data, without an understanding of how those choices impact analysis and problem solving. Further, no tools currently exist to explore the relationships between an LSA model and analysis methods. Our work focuses on how parameter choices impact analysis and problem solving. In this paper, we present LSAView, a system for interactively exploring parameter choices for LSA models. We illustrate the use of LSAView's small multiple views, linked matrix-graph views, and data views to analyze parameter selection and application in the context of graph layout and clustering.

Protein complexes are formed when two or more proteins non-covalently interact to form a larger three dimensional structure with specific biological function. Understanding the composition of such complexes is vital to understanding cell biology at the molecular level. MassVis is a visual analysis tool designed to assist the interpretation of data from a new workflow for detecting the composition of such protein complexes in biological samples. The data generated by the laboratory workflow naturally lends itself to a scatter plot visualization. However, characteristics of this data give rise to some unique aspects not typical of a standard scatter plot. We are able to take the output from tandem mass spectrometry and render the data in such a way that it mimics more traditional two-dimensional gel techniques and at the same time reveals the correlated behavior indicative of protein complexes. By computationally measuring these correlated patterns in the data, membership in putative complexes can be inferred. User interactions are provided to support both an interactive discovery mode as well as an unsupervised clustering of likely complexes. The specific analysis tasks led us to design a unique arrangement of item selection and coordinated detail views in order to simultaneously view different aspects of the selected item.

The Internet traffic challenge required the development of a custom application to analyze internet traffic patterns coupled with building access records. To solve this challenge, the author applied the Prajna Project, an open-source Java toolkit designed to provide various capabilities for visualization, knowledge representation, semantic reasoning, and data fusion. By applying some of the capabilities of Prajna to this challenge, the author could quickly develop a custom application for visual analysis. The author determined that he could solve some of the analytical components of this challenge using automated reasoning techniques. Prajna includes interfaces to incorporate automated reasoners into visual applications. By blending the automated reasoning processes with visual analysis, the author could design a flexible, useful application to solve this challenge.

Discovering and extracting linear trends and correlations in datasets is very important for analysts to understand multivariate phenomena. However, current widely used multivariate visualization techniques, such as parallel coordinates and scatterplot matrices, fail to reveal and illustrate such linear relationships intuitively, especially when more than 3 variables are involved or multiple trends coexist in the dataset. We present a novel multivariate model parameter space visualization system that helps analysts discover single and multiple linear patterns and extract subsets of data that fit a model well. Using this system, analysts are able to explore and navigate in model parameter space, interactively select and tune patterns, and refine the model for accuracy using computational techniques. We build connections between model space and data space visually, allowing analysts to employ their domain knowledge during exploration to better interpret the patterns they discover and their validity. Case studies with real datasets are used to investigate the effectiveness of the visualizations.

Cytoscape is a popular open source tool for biologists to visualize interaction networks. We find that it offers most of the desired functionality for visual analytics on graph data to guide us in the identification of the underlying social structure. We demonstrate its utility in the identification of the social structure in the VAST 2009 Flitter Mini Challenge.