System Advancements at the Monastery

Interpreting data is at the heart of security. Transforming large, often streaming data sets, e-mails, images, numbers and sounds into a form that better supports analytic reasoning continue to become more important as organization have to deal with an ever increasing amount of varied data. While there are many folks doing fascinating work, today’s posts will highlight some of the work being done by researchers as part of the NSF’s Foundations of Data Analysis and Visual Analytics (FODAVA).

Background

In March 2004, DHS established the National Visualization and Analytics Center (NVAC) . The center is led by the Department of Energy’s (DOE) Pacific Northwest National Laboratory (PNNL) in Richland, Washington. NVAC is tasked with providing “scientific guidance and coordination for the research and development of new tools and methods that Homeland Security has identified as required for managing, visually representing, and analyzing enormous amounts of diverse data and information.”

There are five Regional Visualization and Analytics Centers (RVACs) led by Penn State University, Purdue University, Stanford University, the University of North Carolina at Charlotte and Georgia Institute of Technology, and the University of Washington.

Penn State develops visualization tools designed to extract and safely store pertinent information, such as place and time, from a variety of data formats which can help analysts anticipate, prevent and respond to major events. Purdue and IUSM focus is on three homeland security areas – intelligence analysis; emergency planning and response; and healthcare monitoring and management. Stanford performs research on network traffic analysis for intrusion detection; cognitive and perceptual principles supporting reasoning with space and time; and methods to support exploratory analysis of graphs in relational databases. UNC Charlotte and Georgia Tech develop techniques and tools to assist homeland security analysts and then combine the tools in an artificial analytic reasoning system. UW established a Pacific Rim regional center, which includes experts from UW, British Columbia, Australia, New Zealand and Hawaii working together to conduct their analysis, a process known as “collaborative visual analytics.”

FODAVA

In July 2007, DHS announced a partnership between NVAC and National Science Foundation (NSF) to conduct a joint research program in data and visual analytics. The Georgia Institute of Technology was tasked with leading and coordinating the initiative. The NSF’s project, known as the FODAVA, was established to build a community of researchers consisting of Cornell University, Duke University, Northwestern University, Standford University, University of California, Davis, University of Illinois at Chicago, University of Illinois at Urbana-Champaign, University of Michigan. Research is performed in massive data analysis and visual analytics through such areas as machine learning, numeric and geometric computing, optimization, computation statistics, and information visualization.

Sample Work

For an understanding of the type of research being done, take a moment to examine one sample project, Jigsaw. The project is intended to help analysts better assess, analyze and make sense of large document collections. To quote the Jigsaw site description:

Jigsaw is “a visual analytics system to help analysts better assess, analyze, and make sense of such document collections. Our specific objective is to help analysts reach more timely and accurate understandings of the larger stories embedded throughout textual reports. Jigsaw provides a collection of visualizations that each portray different aspects of the documents. We particularly focus on presenting the identifiable important entities (people, places, organizations, etc.) and their direct or indirect connections. Textual processing extracts the important entities from the documents and then the visualizations help an analyst to explore the relationships and connections among the entities.

There is even a tutorial-style video that shows a number of short segments about the different views in Jigsaw. mov (11 MB)

A few additional projects developed by the Information Interfaces Group, an HCI research group in the GVU Center at Georgia Tech, include:

	Jigsaw Using visualization and visual analytics to help analysis and sensemaking on document collections.
	Information Visualization and Visual Analytics Helping people understand and analyze data through interactive visualization techniques and systems.
	Imprint Empowering workers to use visualizations of printer data as a basis for conversation and reflection.
	SellTrend Enabling real-time awareness and exploratory analysis of temporal, categorical event transactions.
	The Buzz Supporting end-user mashup creation and content aggregation onto photo and text collages.
	InfoCanvas Developing Information Art: Virtual paintings that peripherally convey information to people.
	FundExplorer Aiding equity investors with mutual fund portfolio diversification through the use of Context Treemaps.
	Dust & Magnet Assisting people understand multivariate data sets using a magnet metaphor-based visualization.
	Sports Visualization Promoting advanced statistical analysis in sports through the use of information visualization.
	SunBurst Developing circular, space-filling visualizations for depicting information hierarchies and trees.
	Software Visualization Helping people understand software through visualization of programs, data structures, algorithms, and executions.

Below are some of the presentations done for the FODAVA Distinguished Lecture Series:

• Alan Turner (US Government), "Mathematical Foundations as a Key Enabler of Agile Human Performance in Visual Analytics Environments". [pdf]
• William S. Cleveland (Shanti S. Gupta Professor of Statistics and Professor of Computer Science, Purdue University), "The Disappearing Second Derivative of Quadratics: Perceptual, Mathematical, and Statistical Properties of Judging Dependence on Visual Displays". [pdf]
• Joseph Kielman (Science and Technology Directorate, Department of Homeland Security), "Visual Analytics – Past, Present, and Future". [pdf]
• Alexey Chervonenkis (Russian Academy of Science and Royal Holloway University of London), "Model Complexity Optimization".
• Vladimir Vapnik (NEC Laboratories, Columbia University, and Royal Holloway University of London), "Learning with Teacher: Learning Using Hidden Information". [pdf]

Final Thoughts

Visualization is not the solution for every security problem. The work being done by researchers can often seem impractical to those in operations, where unfortunately few of us have high performance supercomputers at our disposal. Still, it is similar to why car manufacturers will take part in building race cars. When you push the envelop, the knowledge learned just might be applicable to those fighting the good fight in organizations across the planet. Watching the work being done at research projects like FODAVA provide a view of interesting possibilities for the future. In combination with existing tools, visualization tools promises to help explore data, discover insights, and provide a way to effectively communicate results. Visualization is a most interesting field offering intriguing possibilities.