Subscribe Bookmark


Jun 13, 2014

Visual Analytics and JMP

I'm at SAS Global Forum this week in Seattle, where I'm doing a couple of Super Demos on Visual Analytics and presenting a data visualization paper. In preparation for the conference and my paper, I looked into the term "Visual Analytics."

In 2004, the US Department of Homeland Security chartered the National Visualization and Analytics Center™ (NVAC™) with the goal of helping to counter future terrorist attacks in the United States and around the globe. A major objective was to define “Visual Analytics.” The intent was to provide a definition and requirements to vendors so that products would be developed or refined to allow for massive amounts to data to be digested and decisions could be made quickly and effectively under emergency conditions. The NVAC defined Visual Analytics as “the science of analytical reasoning facilitated by interactive visual interfaces” in a report titled Illuminating the Path by James J. Thomas and Kristin A. Cook.

The components of Visual Analytics are interactive visualizations using advanced analytics that create statistical models in support of the visuals. These components combine visualization techniques and analysis techniques, that when used together allows analysts to synthesize large amounts of data, detect trends and patterns, and help discover important and unexpected content. This, in turn, facilitates high-quality human judgment with a smaller investment of the analyst’s time.

“Visual analytics seeks to marry techniques from information visualization with techniques from computational transformation and analysis of data. The computer finds patterns in the information and organizes it in ways that are meant to be revealing to the analyst. The analyst supplies his or her knowledge in ways that help the computer refine and organize information more appropriately. Working together, they are much more powerful than each one working separately,” – according to Illuminating the Path.

After reading the 175-page document, I couldn’t help observing that anyone with data would benefit from the type of application the NVAC defined, whether in crisis mode or not. I wish the authors had come to talk to us at JMP. We would have saved them a lot of time (and pages).

Much of what the authors discuss JMP has to offer. JMP allows for ad-hoc exploration of data, using cutting-edge visuals that encourage the user to interact with. It has world-class analytics that further digest data and locate possible root causes and their effects. Its simulation features allow users to see potential outcomes and test their decisions before they have to be implemented. Their document is a good validation of what we’ve done so far, and raised the bar for what we strive to attain in the future.