Guest Editor's Introduction: Special Section on the IEEE Symposium on Visual Analytics Science and Technology (VAST)
Issue No. 02 - March/April (2010 vol. 16)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TVCG.2010.16
Visual Analytics combines techniques from visualization and data analysis with highly interactive user interfaces in order to provide better insight into the ever growing data collections. Although still a young and emerging field, Visual Analytics has already attracted a lot of interest from the research community during the last five years. IEEE Transactions on Visualization and Computer Graphics (TVCG) acknowledged the relevance of this new area with a Call for Papers for a Special Section on Visual Analytics as early as 2005. Thirteen papers were finally published in the November/December 2006 issue of TVCG, and I encourage the readers of this special section to browse this remarkable collection of high-quality papers which helped shape the field. Also in 2006, the IEEE Symposium on Visual Analytics Science and Technology (VAST) had a good start bringing together researchers in this area during what is now called the IEEE VisWeek. This symposium clearly consolidated Visual Analytics research with respect to scope and quality so that TVCG decided to invite some of the best conference papers to submit extended versions which meet the standards of an archival journal. When IEEE VAST took place for the third time in Columbus, Ohio, from 21-23 October 2008, I had the pleasure of selecting the three best papers together with some senior members of the IEEE VAST 2008 International Program Committee. I am happy to report that all invited papers met TVCG's strict 30 percent new material rule and they were finally accepted for publication after one revision cycle.
The first paper is "Principles and Tools for Collaborative Entity-Based Intelligence Analysis" by Eric Bier, Stuart Card, and John Bodnar. It won the best paper award at IEEE VAST 2008. This paper is based on a conceptual analysis of the cognitive tasks performance by intelligence analysts with implications for a broader range of knowledge workers. The sensemaking model that the authors have previously described is a "gold standard" for cognitive task analysis, and it forms the basis for the current implementation. Key aspects of the design include an evidence notebook optimized for organizing entities, information structures that can be collapsed and expanded, visualization of evidence that emphasizes events and documents, and a notification system. The analysis of the system with respect to collaboration using actual intelligence analysts generating a collection of qualitative measures is equally impressive.
The second paper is "Cross-Filtered Views for Multidimensional Visual Analysis" by Chris Weaver. This paper provides a unified model of linked interactions between visualizations of a data set. It describes a method for interactively expressing sequences of multidimensional set queries by cross-filtering data values across pairs of views, and design strategies for constructing coordinated multiple view interfaces for cross-filtered visual analysis of multidimensional data sets. It encapsulates many of the common concepts that have evolved in the field into a nice structure that proves quite powerful on the various given examples. It is an important contribution to the field of Visual Analytics, as it provides a sort of glue for connecting diverse views of the data.
The third paper is "A Visual Analytics Approach to Understanding Spatiotemporal Hotspots" by Ross Maciejewski, Stephen Rudolph, Ryan Hafen, Ahmad Abusalah, Mourad Ouzzani, William Cleveland, Shaun Grannis, and David Ebert. This paper makes a significant contribution to Visual Analytics in that it integrates interactive visualization with careful statistical analyses of spatio-temporal aspects. The authors describe a system which allows the exploration of data aberrations and hotspots through filtering interaction on combined views. Statistical data models and alert detection algorithms can be used to disambiguate hotspot behavior on multiple geo-spatiotemporal data sets.
I hope that this special section gives the readers of TVCG insight into some of the exciting developments within Visual Analytics and encourages them to submit their own best work in this area to TVCG.
For information on obtaining reprints of this article, please send e-mail to: firstname.lastname@example.org.