From Wikipedia, the free encyclopedia Visual analytics
Visual analytics
Visual analytics is an outgrowth of the fields information and machine, amplifying human cognitive capabilities in
visualization and scientific visualization, that focuses on six basic ways:[2][5]
analytical reasoning facilitated by interactive visual in- 1. by increasing cognitive resources, such as by using a
terfaces. [1] visual resource to expand human working memory,
2. by reducing search, such as by representing a large
Overview amount of data in a small space,
3. by enhancing the recognition of patterns, such as
Visual analytics is "the science of analytical reasoning when information is organized in space by its time
facilitated by visual interactive interfaces." [2] It can at- relationships,
tack certain problems whose size, complexity, and need 4. by supporting the easy perceptual inference of
for closely coupled human and machine analysis may relationships that are otherwise more difficult to
make them otherwise intractable.[3] Visual analytics ad- induce,
vances science and technology developments in analyti- 5. by perceptual monitoring of a large number of
cal reasoning, interaction, data transformations and rep- potential events, and
resentations for computation and visualization, analytic 6. by providing a manipulable medium that, unlike
reporting, and technology transition. [4] As a research static diagrams, enables the exploration of a space of
agenda, visual analytics brings together several scientific parameter values.
and technical communities from computer science, in- These capabilities of information visualization, combined
formation visualization, cognitive and perceptual sci- with computational data analysis, can be applied to ana-
ences, interactive design, graphic design, and social sci- lytic reasoning to support the sense-making process.
ences.
Visual analytics integrates new computational and
theory-based tools with innovative interactive tech-
Topics
niques and visual representations to enable human-in-
formation discourse. The design of the tools and tech- Scope
niques is based on cognitive, design, and perceptual prin- Visual analytics is a multidisciplinary field that includes
ciples. This science of analytical reasoning provides the the following focus areas:[2]
reasoning framework upon which one can build both • Analytical reasoning techniques that enable users to
strategic and tactical visual analytics technologies for obtain deep insights that directly support
threat analysis, prevention, and response. Analytical rea- assessment, planning, and decision making
soning is central to the analyst’s task of applying human • Data representations and transformations that
judgments to reach conclusions from a combination of convert all types of conflicting and dynamic data in
evidence and assumptions.[2] ways that support visualization and analysis
Visual analytics has some overlapping goals and tech- • Techniques to support production, presentation, and
niques with information visualization and scientific vi- dissemination of the results of an analysis to
sualization. There is currently no clear consensus on the communicate information in the appropriate
boundaries between these fields, but broadly speaking context to a variety of audiences.
the three areas can be distinguished as follows: • Visual representations and interaction techniques
• Scientific visualization deals with data that has a that take advantage of the human eye’s broad
natural geometric structure (e.g., MRI data, wind bandwidth pathway into the mind to allow users to
flows). see, explore, and understand large amounts of
• Information visualization handles abstract data information at once
structures such as trees or graphs.
• Visual analytics is especially concerned with Analytical reasoning techniques
sensemaking and reasoning. Analytical reasoning techniques are the method by
Visual analytics seeks to marry techniques from infor- which users obtain deep insights that directly support
mation visualization with techniques from computation- situation assessment, planning, and decision making.
al transformation and analysis of data. Information visu- Visual analytics must facilitate high-quality human judg-
alization forms part of the direct interface between user ment with a limited investment of the analysts’ time.
1
From Wikipedia, the free encyclopedia Visual analytics
Visual analytics tools must enable diverse analytical
tasks such as:[2]
Process
• Understanding past and present situations quickly, The input for the data sets used in the visual analytics
as well as the trends and events that have produced process are heterogeneous data sources (i.e., the inter-
current conditions net, newspapers, books, scientific experiments, expert
• Identifying possible alternative futures and their systems). From these rich sources, the data sets S = S1, ...,
warning signs Sm are chosen, whereas each Si , i ? (1, ..., m) consists of at-
• Monitoring current events for emergence of warning tributes Ai1, ..., Aik. The goal or output of the process is in-
signs as well as unexpected events sight I. Insight is either directly obtained from the set of
• Determining indicators of the intent of an action or created visualizations V or through confirmation of hy-
an individual potheses H as the results of automated analysis methods.
• Supporting the decision maker in times of crisis. This formalization of the visual analytics process is illus-
These tasks will be conducted through a combination trated in the following figure. Arrows represent the tran-
of individual and collaborative analysis, often under ex- sitions from one set to another one.
treme time pressure. Visual analytics must enable More formal the visual analytics process is a transfor-
hypothesis-based and scenario-based analytical tech- mation F : S → I, whereas F is a concatenation of functions
niques, providing support for the analyst to reason based f ? {DW, VX, HY, UZ} defined as follows:
on the available evidence.[2] DW describes the basic data pre-processing function-
ality with DW : S → S and W ? {T, C, SL, I} including data
Data representations transformation functions DT, data cleaning functions DC,
Data representations are structured forms suitable for data selection functions DSL and data integration func-
computer-based transformations. These structures must tions DI that are needed to make analysis functions ap-
exist in the original data or be derivable from the data plicable to the data set.
themselves. They must retain the information and VW, W ? {S, H} symbolizes the visualization functions,
knowledge content and the related context within the which are either functions visualizing data VS : S → V or
original data to the greatest degree possible. The struc-
functions visualizing hypotheses VH : H → V.
tures of underlying data representations are generally
HY, Y ? {S, V} represents the hypotheses generation
neither accessible nor intuitive to the user of the visual
process. We distinguish between functions that generate
analytics tool. They are frequently more complex in na-
hyphotheses from data HS : S → H and functions that gen-
ture than the original data and are not necessarily small-
er in size than the original data. The structures of the da- erate hypotheses from visualizations HV : V → H.
ta representations may contain hundreds or thousands Moreover, user interactions UZ, Z ? {V, H, CV, CH} are
of dimensions and be unintelligible to a person, but they an integral part of the visual analytics process. User in-
must be transformable into lower-dimensional represen- teractions can either effect only visualizations UV : V → V
tations for visualization and analysis.[2] (i.e., selecting or zooming), or can effect only hypotheses
UH : H → H by generating a new hypotheses from given
Theories of visualization ones. Furthermore, insight can be concluded from visual-
Theories of visualization are:[3] izations UCV : V → I or from hypotheses UCH : H → I.
• "Semiology of Graphics" in 1967 written by Jacques The typical data pre-processing applying data clean-
Bertin e ing, data integration and data transformation functions
• "Languages of Art" from 1977 by Nelson Goodman is defined as DP = DT(DI(DC(S1, ..., Sn))). After the pre-pro-
• Jock D. Mackinlay’s "Automated design of optimal cessing step either automated analysis methods HS = {fs1,
visualization" (APT) from 1986, and ..., fsq} (i.e., statistics, data mining, etc.) or visualization
• Leland Wilkinson’s "Grammar of Graphics" from methods VS : S → V, VS = {fv1, ..., fvs} are applied to the
1998, data, in order to reveal patterns as shown in the figure
above.[6]
Visual representations In general the following paradigm is used to process
Visual representations translate data into a visible form the data:
that highlights important features, including commonal- Analyse First – Show the Important – Zoom, Filter and Analyse
ities and anomalies. These visual representations make Further – Details on Demand[7]
it easy for users to perceive salient aspects of their data
quickly. Augmenting the cognitive reasoning process
with perceptual reasoning through visual representa-
tions permits the analytical reasoning process to become
faster and more focused.[2]
2
From Wikipedia, the free encyclopedia Visual analytics
Flow As First Class Citizens in [4] Kielman, J. and Thomas, J. (Guest Eds.) (2009).
"Special Issue: Foundations and Frontiers of Visual
Computing Analytics". in: Information Visualization, Volume 8,
Number 4, Winter 2009 Page(s): 239-314.
It is only in recent times that flows have been represent-
[5] Stuart Card, J.D. Mackinlay, and Ben Shneiderman
ed as first class data items to build the web technology e.g
(1999). "Readings in Information Visualization:
JSF flows or Spring Web Flows. Also the sense of flow and
Using Vision to Think". Morgan Kaufmann
sense of focus has been proposed as two different senses
Publishers, San Francisco.
in our brain, This is an example of how the science of vi-
[6] Daniel A. Keim, Florian Mansmann, Jörn
sual analytics can bring sense and richness in our under-
Schneidewind, Jim Thomas, and Hartmut Ziegler
standing and control of complex process in our computa-
(2008). "Visual Analytics: Scope and Challenges"
tion and the processes in our brain.
[7] Keim D. A, Mansmann F, Schneidewind J, Thomas J,
Ziegler H: Visual analytics: Scope and challenges.
See also Visual Data Mining: 2008, S. 82.
• Argument mapping
• Business Decision Mapping Further reading
• Cartography
• Boris Kovalerchuk and James Schwing (2004). Visual
• Computational visualistics
and Spatial Analysis: Advances in Data Mining, Reasoning,
• Critical thinking
and Problem Soving
• Decision making
• Guoping Qiu (2007). Advances in Visual Information
• Diagrammatic reasoning
Systems: 9th International Conference (VISUAL).
• Geovisualization
• IEEE, Inc. Staff (2007). Visual Analytics Science and
• Google Analytics
Technology (VAST), A Symposium of the IEEE 2007.
• Social network analysis software
• May Yuan, Kathleen and Stewart Hornsby (2007).
• Software visualization
Computation and Visualization for Understanding
• Starlight Information Visualization System
Dynamics in Geographic Domains.
• Text analytics
• Traffic analysis
• Visual reasoning External links
• Wicked problem • VisMaster Visual Analytics – Mastering the
Related scientists Information Age
• Cecilia R. Aragon • SPP - Scalable Visual Analytics
• Robert E. Horn • Visual Analytics a course by Robert Kosara, 2007.
• Daniel A. Keim • IEEE Visual Analytics Science and Technology (VAST)
• Theresa-Marie Rhyne Symposium
• Lawrence J. Rosenblum • National Visualization and Analytics Center (NVAC)
• John Stasko • Visual Analytics Digital Library (VADL)
• Jim_Thomas_(visualization) • GeoAnalytics.net - GeoSpatial Visual Analytics, ICA
• William Ribarsky Commission on GeoVizualisation
• International Cartographic Association (ICA), the
References world body for mapping and GIScience professionals
• flowingdata.com, visualization and statistics blog
[1] Pak Chung Wong and J. Thomas (2004). "Visual
• visual-analytics.org, visual analytics blog
Analytics". in: IEEE Computer Graphics and
• Middlesex University Interaction Design Centre
Applications, Volume 24, Issue 5, Sept.-Oct. 2004
weblog, visual analytics blog and resources
Page(s): 20–21.
• Visual Analytics :Focusing on web pages design and
[2] ^ James J. Thomas and Kristin A. Cook (Ed.) (2005).
content as a visual analytics of semantics and human
Illuminating the Path: The R&D Agenda for Visual
understanding, one of the pioneer users of the
Analytics National Visualization and Analytics
phrase Visual Analytics
Center. p.4.
• Understanding Link Analysis, link and visual analysis
[3] ^ Robert Kosara (2007). Visual Analytics. ITCS 4122/
resource
5122, Fall 2007. Retrieved 28 june 2008.
Retrieved from "http://en.wikipedia.org/w/index.php?title=Visual_analytics&oldid=457665568"
3
From Wikipedia, the free encyclopedia Visual analytics
Categories:
• Computational science
• Computer graphics
• Infographics
• Visualization (graphic)
• Scientific modeling
• Cartography
This page was last modified on 27 October 2011 at 15:00. Text is available under the Creative Commons Attribution-
ShareAlike License; additional terms may apply. See Terms of use for details. Wikipedia® is a registered trademark of
the Wikimedia Foundation, Inc., a non-profit organization.Contact us
Privacy policy About Wikipedia Disclaimers Mobile view
4