Graph and Network
Visualization
John Stasko
Georgia Institute of Technology
Connections
• Connections throughout our lives and the
world
− Circle of friends
− Delta’s flight plans
−…
• Model connected set as a Graph
InfoVis 2
What is a Graph?
• Vertices (nodes)
connected by
• Edges (links) Adjacency list
1: 2
2: 1, 3
1 2 3 3: 2 1
1 0 1 0
2 1 0 1 2 3
3 0 1 0
Drawing
Adjacency matrix
InfoVis 3
Graph Terminology
• Graphs can have cycles
• Graph edges can be directed or
undirected
• The degree of a vertex is the number of
edges connected to it
− In-degree and out-degree for directed graphs
• Graph edges can have values (weights)
on them (nominal, ordinal or quantitative)
InfoVis 4
Trees are Different
• Subcase of general graph
• No cycles
• Typically directed edges
• Special designated root vertex
InfoVis 5
Graph Uses
• In information visualization, any number
of data sets can be modeled as a graph
− US telephone system
− World Wide Web
− Distribution network for on-line retailer
− Call graph of a large software system
− Semantic map in an AI algorithm
− Set of connected friends
InfoVis 6
Graph Visualization Problems
• Graph layout and positioning
− Make a concrete rendering of abstract graph
• Scale
− Not too much of a problem for small graphs,
but large ones are much tougher
• Navigation/Interaction
− How to support user changing focus and
moving around the graph
InfoVis 7
Layout Algorithms
• Entire research community’s focus
• Good references:
− Tutorial (talk slides)
www.cs.brown.edu/people/rt/papers/gd-tutorial/gd-constraints.pdf
− G. diBattista, P. Eades, R. Tamassia, and I.
Tollis, Graph Drawing: Algorithms for the
Visualization of Graphs, Prentice Hall, 1999.
InfoVis 8
General GD Information
• Good web links
− www.cs.brown.edu/people/rt/gd.html
− www.research.att.com/sw/tools/graphviz/
− rw4.cs.uni-sb.de/users/sander/html/gstools.html
InfoVis 9
Graph Drawing Conference
InfoVis 10
Vertex Issues
• Shape
• Color
• Size
• Location
• Label
InfoVis 11
Edge Issues
• Color
• Size
• Label
• Form
− Polyline, straight line, orthogonal, grid,
curved, planar, upward/downward, ...
InfoVis 12
Aesthetic Considerations
• Crossings -- minimize towards planar
• Total Edge Length -- minimize towards proper
scale
• Area -- minimize towards efficiency
• Maximum Edge Length -- minimize longest
edge
• Uniform Edge Lengths -- minimize variances
• Total Bends -- minimize orthogonal towards
straight-line
InfoVis 13
Which Matters?
• Various studies examined which of the
aesthetic factors matter most and/or what
kinds of layout/vis techniques look best
− Purchase, Graph Drawing ’97
− Ware et al, Info Vis 1(2), June ’02
− Ghoniem et al, Info Vis 4(2), Summer ’05
−…
• Results mixed: Edge crossings seem
important
InfoVis 14
Layout Heuristics
• Layout algorithms can be
− planar
− grid-based
− orthogonal
− curved lines
− hierarchies
− circular
− ...
InfoVis 15
Types of Layout Algorithms
From:
P. Mutzel, et al
Graph Drawing ‘97
InfoVis 16
Layout Examples
• Cool java applet
http://java.sun.com/applets/jdk/1.2/demo/applets/GraphLayout/example1.html
• Examples of dynamic graph layout
algorithms
InfoVis 17
Scale Challenge
• May run out of space for vertices and
edges (turns into “ball of string”)
• Can really slow down algorithm
• Often use clustering to help
− Extract highly connected sets of vertices
− Collapse some vertices together
InfoVis 18
Navigation/Interaction Issues
• How do we allow a user to query, visit, or
move around a graph?
• Changing focus may entail a different
rendering
InfoVis 19
Graph/Network Visualization
• One of the oldest and most studied areas
of information visualization
• Good overview
− Herman et al, IEEE TVCG ’00
InfoVis 20
Examples www.visualcomplexity.com
InfoVis 21
Graph Uses
• Facilitate understanding of complex socio-
economic patterns
• Social Science visualization gallery (Lothar
Krempel):
− http://www.mpi-fg-koeln.mpg.de/~lk/netvis.html
• Next slides: Krempel & Plumper’s study
of World Trade between OECD countries,
1981 and 1992
InfoVis 22
http://www.mpi-fg-koeln.mpg.de/~lk/netvis/trade/WorldTrade.html 1981
1992
Social Network Visualization
• Social Network Analysis (Linton Freeman)
− http://www.sfu.ca/~insna
InfoVis 25
People connections
InfoVis
Charles Isbell, Cobot 26
Vizster
Saw in Social Visualization lecture
InfoVis 27
SocialAction
Perer & Shneiderman
InfoVis InfoVis ‘06 28
Graph Uses
• Facilitate understanding of network flows,
relations
• Even information with a ‘geographical’
content can best appear as a ‘network’ rail
maps
InfoVis 29
3 Subway Diagrams
• Geographic landmarks largely suppressed
on maps, except water (rivers in Paris,
London) and asphalt (highways in Atlanta)
− Rather fitting, no?
• These are more graphs than maps!
InfoVis 33
Case Study
• SeeNet
− Visualizing network data (phone traffic)
R. Becker, S. Eick and A. Wilks
AT&T
InfoVis 34
Domain
• AT&T long distance phone network
− 110 Nodes (switches)
Geographical location
− Connected by 12,000 links
Directed, almost completely connected
• Data every 5 minutes
• EARTHQUAKE!!!
− Oct. 17, 1989
InfoVis 35
Questions
• Where are the overloads?
• Which links are carrying most traffic?
• Was there network damage?
• Is there underutilized capacity?
• Are calls getting in to affected area or are
there bottlenecks?
• Is overload increasing or decreasing?
InfoVis 36
Edge Drawing Strategies
116
Label
Thickness
Color
116
29
Directed
InfoVis 37
Problems
• Too many lines!
− Occlusion
− Long lines become “more important”
− Can’t see what happens in Midwest
• Solutions
− Use half/half technique out/out
− Draw most important last
− Use thickness & color for traffic
InfoVis 38
Earthquake data
More Help
• Shorten all lines so as to de-emphasize
transcontinental links
InfoVis 40
Case Study
• NicheWorks
− Interactive Visualization of Very Large Graphs
Graham Wills
Lucent (at time)
InfoVis 42
Big Graphs
• 20,000 - 1,000,000 Nodes
• Works well with 50,000
• Projects
− Software Engineering
− Web site analysis
− Large database correlation
− Telephone fraud detection
InfoVis 43
Features
• Typical interactive operations
• Sophisticated graph layout algorithm
− 3 Layouts
Circular
Hexagonal
Tree
− 3 Incremental Algorithms
Steepest Descent
Swapping
Repelling
InfoVis 44
Web Site Example
Circle layout Hexagonal layout Tree layout
InfoVis 45
Interface
InfoVis 46
Interface
InfoVis 47
Phone Fraud Example
Shown are
people calling
that country
Length of
edge is
duration of
call
40,000 calls
35,000 callers
InfoVis 48
Fraud Example
Filtering for people
who made multiple
calls and spent a
significant amount
of time on the phone
Playing with parameters
like these is important
because fraudsters
know how to evade
Note the two people
calling Israel and Jordan
InfoVis 49
Fraud Example
Zooming in, we notice they have
similar calling patterns and numbers
(likely part of same operation)
Illegal to call between Israel and
Jordan at the time, so fraudsters
set up rented apts in US and charge
Israeli and Jordanian business people
for 3rd party calling
When bills came to US, they would
ignore and move on
InfoVis 50
More Neat Stuff
• http://willsfamily.org/gwills/
• Lots of interesting application areas
• More details on NicheWorks
InfoVis 51
Other Applications
• Email
• How would you visualize all email traffic in
your department between pairs of people?
• Solutions???
InfoVis 52
Solutions
• Put everyone on circle, lines between
− Color or thicken line to indicate magnitude
• Use spring/tension model
− People who send a lot to each other are
drawn close together
− Shows clusters of communications
InfoVis 53
More Email
• How about visualizing internet traffic?
InfoVis 54
Byte traffic into the ANS/NSFnet T3 backbone for the month of November, 1993
http://www.ncsa.uiuc.edu/SCMS/DigLib/text/technology/Visualization-Study-NSFNET-Cox.html
Inbound traffic measured in billions of bytes on the NSFNET T1 backbone for September 1991
Linux kernel
http://perso.wanadoo.fr/pascal.brisset/kernel3d/kernel3d.html
InfoVis 57
TouchGraph
www.touchgraph.com
InfoVis 58
Focus of Graph
• Particular node may be focus, often
placed in center for circular layout
• How does one build an interactive system
that allows changes in focus?
− Use animation
− Intuition about changes not always right
InfoVis 59
Focus Change Animation
Straight linear interpolation
of focus changes not as
appealing as changes along
polar coordinates
Yee, Fisher, Dhamija, Hearst
InfoVis ‘01
Video
InfoVis 60
MoireGraphs
• Combine a focus + context radial graph
layout with a variety of interaction
techniques
• Useful when visual information such as
images is also present at each node and
must be displayed
Jankun-Kelly & Ma
InfoVis ‘03
InfoVis 61
Navigation and interaction…
Video
InfoVis 62
Alternative Approaches
• Recently, researchers have been exploring
approaches other than the traditional
node-link views…
InfoVis 63
Adjacency Matrices
MatrixExplorer
Henry & Fekete
InfoVis InfoVis ‘06 64
PivotGraph
Cluster nodes on an attribute value
Wattenberg
Show links between sets
CHI ‘06
InfoVis 65
Semantic Substrates
Separate nodes by attribute
Use dynamic query
Shneiderman & Aris capabilities
InfoVis InfoVis ‘06 66
More Resources
• Network visualization resources
− http://www.caida.org/projects/internetatlas/viz/
• Good article on graph layout
− http://www.csi.uottawa.ca/ordal/papers/sander/main.html
InfoVis 67
More to Come...
• Topic of WWW/InfoSphere (next lecture)
will touch on graphs and networks too
• Lots of example visualizations
InfoVis 68
End
• References
− Spence and CMS texts
− All referred to papers and web sites
− Dagon and Leahy, F ‘99 slides
InfoVis 69
