|
Improvise Videos
Improvise is an information visualization builder and browser
that has been used to explore census data, ham radio communications,
historic hotel guest visitation patterns, election results,
hydrographics, MP3 music collections, the chemical elements,
and even the interactive structure of its own visualizations
in situ. These videos demostrate Improvise in use:
Visual Analysis of Historic Hotel Visitation Patterns
(96 MB)
Space-Time
Visualization on Multiple Displays (250 MB)
Cinegraph:
2007 Infovis winning entry (68 MB)
Avian Flu Viewer
This narrated video shows our tool for exploring Avian Flu
research: Avian
Flu Viewer
Visual Inquiry Toolkit
The Visual Inquiry Toolkit integrates visual, computational,
and cartographic methods to enable human knowledge and judgment
to be coupled productively with computational methods for
incrementally searching patterns. It was initially developed
and applied to dataset on U.S. industry and sales, which was
provided as the benchmark data set for the 2005 IEEE Information
Visualization Contest. Our Visual Inquiry Toolkit entry took
first place. We are now generalizing the methods for application
to cancer incidence and covariate data. Please check back
for updates in the summer of 2006.
Read a summary paper of the winning entry here
(pdf file)
Watch a video demonstration of the
winning entry here
(42 MB)
NeoCITIES
Viewed as both an update and extension of the original CITIES
task, NeoCITIES is used to test team collaborative decision-making
processes, knowledge acquisition and knowledge management
within a command, control and computer-mediated communication
(C4) environment. As an adaptable problem interface, NeoCITIES
is designed to allow researchers to closely examine team behaviors,
identify patterns of response to time-stressed situations,
and monitor the performance outcomes of semi-autonomous, spatially
distributed, decision-making teams. For more information and
a video about NeoCITIES, visit the MINDS
website.
Dialogue-Assisted Visual Environment for Geoinformation
(DAVE_G)
Current computing systems restrict human-computer interaction
to one mode at a time and are designed with an assumption
that use will be by individuals (rather than groups), directing
(rather than interacting with) the system. To support the
ways in which humans work and interact, a new paradigm for
computing is required that is multimodal, rather than unimodal,
collaborative, rather than personal, and dialogue-enabled,
rather than unidirectional. We outline a conceptual approach
toward natural, multimodal, multiuser dialogue-enabled interfaces
to geographic information systems (DAVE_G) that make use of
large-screen displays and virtual environment technology.
Watch a demonstration of DAVE_G
and a Multimodal
Tablet GIS demonstration
Building ESTAT in GeoVISTA Studio:
How
to Build ESTAT In GeoVISTA Studio
Sample
Dataset
Building ESTAT in GeoVISTA
Studio (.avi video clip - if you have trouble,
try downloading this)
Description: This clip illustrates the steps taken to build
ESTAT in GeoVISTA Studio. The process starts by adding 6 Java
Beans to the Studio design box: one bivariate map, one scatterplot,
two parallel coordinate plots (one used as a PCP and the other
used as a time-series graph), one data loader, and one coordinator.
Then, beans are wired together so that all of the graphic
components support coordinated user action and one of the
PCPs works as a time series plot (by being linked to the data
loader's time series import hook, rather than directly to
the coordinator).
Video Tutorials - Using ESTAT to
Explore Cancer Data:
| Title |
Format |
| ESTAT_map-scatterplot-fish |
.avi (trouble? try this) |
| Description: This clip focuses on
exploration of bivariate data relationships using the
bivariate map and scatterplot tools. The clip begins by
illustrating mouse-over ID of places in the scatterplot,
then illustrates linked brushing from the scatterplot
to the map and PCP, then from the map to the other components.
Then, the excentric labeling component is illustrated
(this is a free component obtained from http://www.cs.umd.edu/hcil/excentric/).
Finally, a fisheye lens applied to the map is demonstrated.
This component is from an open source tookkit by Jean-Daniel
Fekete called The InfoVis Toolkit, available from: http://www.lri.fr/~fekete/ |
| ESTAT_stomach-benzine_extreme |
.avi (trouble? try this) |
| Description: This clip emphasizes
the use of brushing on the PCP. The variables depicted
are mercury and benzene emissions, stomach cancer, and
per capita income. The first scene shows that Los Angeles
county has the highest benzene emissions (and third highest
mercury emissions). The interaction demonstrated shows
a two step selection. First, those counties with benzene
emission above the interquartile range are brushed. Then,
all the remaining counties with stomach cancer mortality
rates within or below the interquartile range are removed
from the display. The counties that remain are those that
are both high in stomach cancer mortality and high in
bezine emission. The map illustrates that these counties
are concentrated in the northeast industrial belt and
in the west. The preponderance of dark gray counties indicates
that the counties are also high in mercury emissions. |
| ESTAT_box_state_extremes_Br |
.avi (trouble? try this) |
| Description: This clip is also focused
on the PCP. The initial scene shows that the county with
highest male stomach cancer mortality for the time period
is Petroleum, MN. The rest of the video clip illustrates
the use of box plots, summary lines (one for each state),
and axis focusing and rescaling. The tools are used to
identify states in the U.S. having stomach cancer mortality
rates for men that are (a) lower than the interquartile
range as depicted on the box plot for stomach cancer (mostly
in the southeast) and (b) above the interquartile range
(mostly in the northeast and upper mid-west/great plains).
The map and scatterplot depict the relation between stomach
cancer mortality and mercury emissions aggregated by county.
When the PCP is focused in on the high stomach cancer
mortality counties, the bivariate map illustrates that
those in the northeast are also high in mercury emission
(indicated by the dark gray that represents high values
on both variables) while the mid-west/great plains counties
do not exhibit this relationship. |
| ESTAT_Lung_Cancer_Gender_investigation |
Flash .swf |
| Description: This video demonstration
illustrates the visual analysis of a multivariate dataset
using ESTAT. The dataset explored here is made up of lung
cancer mortality counts and rates and a variety of socioeconomic,
environmental, and educational variables, all aggregated
to the county level across the continental United States.
The demonstration shows how visualization and linked tools
in ESTAT can be used to quickly and broadly to investigate
the dataset for geographic variations in any of the variables.
It then uses, as an example, differences in mortality
by gender to illustrate a deeper multivariate analysis
of the data. The analysis reveals some interesting associations
between male lung cancer mortality and other variables
that generate deeper questions and warrant further analysis
with ESTAT, and with traditional epidemiological studies.
|
|
