Introduction to the Special Issue on Visual Analytics
Daniel A. Keim
Databases, Data Mining and Visualization Group
Department of Computer Science
University of Konstanz
Germany
keim@inf.uni-konstanz.de
Joern Schneidewind
Databases, Data Mining and Visualization Group
Department of Computer Science
University of Konstanz
Germany
jschneidewind@acm.org
The information overload is a well-known phenomenon of
the information age, since due to the progress in computer
power and storage capacity over the last decades, data is
produced at an incredible rate, and our ability to collect
and store these data is increasing at a faster rate than our
ability to analyze it. But, the analysis of these massive,
typically messy and inconsistent, volumes of data is crucial
in many application domains. For decision makers, analysts
or emergency response teams it is an essential task to rapidly
extract relevant information from the flood of data.
Today, a selected number of software tools is employed to
help analysts to organize their information, generate over-
views and explore the information space in order to extract
potentially useful information. Most of these data analysis
systems still rely on interaction metaphors developed more
than a decade ago and it is questionable whether they are
able to meet the demands of the ever-increasing mass of in-
formation. In fact, huge investments in time and money are
often lost, because we still lack the possibilities to properly
interact with the databases.
Visual analytics aims at bridging this gap by employing
more intelligent means in the analysis process. The basic
idea of visual analytics is to visually represent the infor-
mation, allowing the human to directly interact with the
information, to gain insight, to draw conclusions, and to ul-
timately make better decisions. The visual representation of
the information reduces complex cognitive work needed to
perform certain tasks. People may use visual analytics tools
and techniques to synthesize information and derive insight
from massive, dynamic, and often conflicting data by pro-
viding timely, defensible, and understandable assessments.
This special issue presents articles that address interesting,
imp ortant and diverse iss ues in visual analytics research and
practice.
Scope of Visual Analytics
The goal of visual analytics research is to turn the informa-
tion overload into an opportunity. Decision-makers should
be enabled to examine this massive, multi-dimensional, multi-
source, time-varying information stream to make effective
decisions in time-critical situations. For informed decisions,
it is indispensable to include humans in the data analysis
process to combine their flexibility, creativity, and back-
ground knowledge with the enormous storage capacity and
the computational power of today’s computers [1].
The specific advantage of Visual Analytics is that de cision
makers may focus their full cognitive and perceptual capa-
bilities on the analytical process, while allowing them to
apply advanced computational capabilities to augment the
exploration process [5].
Presentation, production,
and dissimilation
Geospatial Analytics
Scientific Analytics
Knowledge Discovery
Data Management &
Knowledge Representation
Cognitive and
Perceptual Science
Interaction
Statistical Analytics
Scope of Visual
Analytics
Information Analytics
Figure 1: The Scope of Visual Analytics
In general, Visual Analytics can be described as “the sci-
ence of analytical reasoning facilitated by interactive visual
interfaces” [4]. To be more precise, Visual Analytics is an
iterative process that involves information gathering, data
preprocessing, knowledge representation, interaction and de-
cision making. The ultimate goal is to gain insight in the
problem at hand which is described by vast amounts of sci-
entific, forensic or business data from heterogeneous sources.
To reach this goal, Visual Analytics combines the strengths
of machines with those of humans [1].
On one hand, methods from data mining, statistics, and
mathematics are the driving force on the automatic analysis
side, while on the other hand human capabilities to p er-
ceive, relate, and conclude turn Visual Analytics into a very
promising field of research.
The Visual Analytics process, as described in [2], aims at
tightly coupling automated analysis methods and interac-
tive visual representations, as described before. Without
the support of automated methods, the visual analysis of
large data sets will become impossible in the future.
Page 3
First publ. in: ACM SIGKDD explorations newsletter 9 (2007), 2, pp. 3-4
Konstanzer Online-Publikations-System (KOPS)
URL: http://www.ub.uni-konstanz.de/kops/volltexte/2008/6852/
URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-opus-68522
As a consequence, we extended the classical way of visu-
ally exploring data sets as defined by the Information Seek-
ing Mantra (“Overview first, Zoom/ Filter, Details on de-
mand”)[3], to the Visual Analytics Mantra [2]:
“Analyze First -
Show the important -
Zoom, Filter and Analyse Further -
Details on Demand”
In the age of massive data sets all three steps of the Infor-
mation Seeking Mantra are difficult to realize. An overview
visualization without losing any interesting pattern or sub-
set is difficult to create, since the amount of pixels of mo dern
data display systems do not keep pace with the increasing
flo od of data. The plenty of information hidden in massive
data sets make it very difficult for humans to understand
the really interesting or relevant information.
In Visual Analytics it is therefore not sufficient to just re-
trieve and display the data using a visual metaphor, it is
rather necessary to support the analyst by analytically fil-
tering the underlying data by its value of interest, but at the
same time providing interaction models which still allow the
user to get any detail of the data on demand.
About the articles
This special issue on Visual Analytics attracted a number
of high quality submissions from Brazil, France, Germany,
Italy, the Netherlands, Switzerland, UK, and the US. Each
submission was assigned to at least two domain experts for
thorough review and evaluation. We selected five submis-
sions that best represent the current state of the art in the
studies and applications of visual analytics.
I. Assent, R. Krieger, E. Mueller, and T. Seidl describe a
subspace clustering visualization that allows users to browse
the entire subspace clustering, to zo om into individual ob-
jects, and to analyze subspace cluster characteristics in-
depth. Browsing of the clustering result is possible through
a novel distance function that reflects the subspace and the
object overlap, respectively. Since the pap er shows how vi-
sual methods can be tighly coupled with automated data
mining techniques it address one of the core issues of visual
analytics.
H. Kang, L. Getoor, and L. Singh investigate the use of vi-
sual analytics techniques for analyzing dynamic group mem-
bership in temporal social networks over time. They present
the C-Group tool, that unlike most network visualization
to ols, which show the group structure within an entire net-
work, or the group membership for a single actor, allows
users to focus their analysis on a pair of individuals. The
authors show how dynamic patterns can be explored in com-
plex networks, which is a valuable contribution for a number
of visual analytics application domains.
The article by D. Yang, Z. Xie, E. A. Rundensteiner, and M.
O. Ward presents an innovative visual analytics framework
for analysis-guided visual exploration of multivariate data.
Their system helps users extract the valuable information
(nuggets) hidden in datasets based on their interests. Vi-
sualization and interaction techniques are designed to help
users observe and organize the extracted nuggets in an in-
tuitive manner and eventually faciliate their sense-making
process. T hus the paper provides a promising combination
of automated and interactive techniques for multi-variate
data analysis.
The work by T. Schreck, T. Tekusova, D. Fellner, and J.
Kohlhammer descibes a novel approach for an application
of visual analytics techniques in the financial sector, in par-
ticular the analysis of financial time-varying indicator data.
The system relies on an unsupervised clustering algorithm
combined with an appropriately designed movement data
visualization technique. Several analytical views on the full
market and specific assets are offered for the user to navi-
gate, to explore, and to analyze. The presentet tool allows
even non-domain experts to quickly get an overview over
asset risk-return patterns.
Finally, G. Andrienko, N. Andrienko, and S. Wrobel focus
on the use of innovative visual analytics techniques to an-
alyze and visualize geospatial patterns. They explore large
volumes of GPS data and prese nt a framwork to effectively
support human analysts in understanding movement behav-
iors and mobility patterns. They demonstrate the synergis-
tic use of automated and visual techniques in case studies
of two real world datasets. The presented ideas are suited
to analyze geo-related information in a number of related
visual analytics scenarios.
Conclusion
The articles presented in this special issue represent some
of the most important topics for further investigation and
exploration, and show that the integration of automated
and visual/interactive techniques, as proposed in the context
of visual analytics, is of great benefit for many application
domains.
Acknowledgments
We are grateful to the reviewers for their timely and con-
structive reviews. Without their help, this special issue
would not be possible. We also thank the authors who re-
sponded to the call-for-papers and submitted their articles
to this special issue.
REFERENCES
[1] D. A. Keim. Information visualization and visual data
mining. IEEE Transactions on Visualization and Com-
puter Graphics (TVCG), 8(1):1–8, January–March 2002.
[2] D. A. Keim, F. Mansmann, J. Schneidewind, and
H. Ziegler:. Challenges in visual data analysis. In IEEE
Information Visualization, London, UK, 2006.
[3] Ben Shneiderman. The eyes have it: A task by data type
taxonomy for information visualizations. In IEEE Visual
Languages, pages 336–343, 1996.
[4] J. Thomas and K. Cook. Illuminating the Path: Research
and Development Agenda for Visual Analytics. IEEE-
Press, 2005.
[5] Pak Chung Wong and Jim Thomas. Visual analytics -
guest editors’ introduction. IEEE Transactions on Com-
puter Graphics and Applications, September/October
2004.
Page 4
