Transcript Thinking Interactively with Visualizations

Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Thinking Interactively with Visualizations
Remco Chang
UNC Charlotte
Charlotte Visualization Center
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Definition of Visual Analytics
• Visual analytics is the science of analytical reasoning
facilitated by interactive visual interfaces [Thomas &
Cook 2005]
• Since 2004, the field has
grown significantly. Aside
from tens to hundreds of
domestic and international
partners, it now has a IEEE
conference (IEEE VAST), an
NSF program (FODAVA), and a
forthcoming IEEE Transactions
journal.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Individually Not Unique
• Data Mining
• Machine
Learning
• Databases
• Information
Retrieval
• etc
Analytical
Reasoning
and
Interaction
Data
Representation
Transformation
Production,
Presentation
Dissemination
• Tech Transfer
• Report Generation
• etc
•
•
•
•
Interaction Design
Cognitive Psychology
Intelligence Analysis
etc.
Visual
Representation
•
•
•
•
InfoVis
SciVis
Graphics
etc
Validation
and
Evaluation
• Quality Assurance
• User studies (HCI)
• etc
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
In Combinations of 2 or 3…
• Data Mining
• Machine
Learning
• Databases
• Information
Retrieval
• etc
Analytical
Reasoning
and
Interaction
Data
Representation
Transformation
Production,
Presentation
Dissemination
Visual
Representation
Validation
and
Evaluation
•
•
•
•
InfoVis
SciVis
Graphics
etc
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
In Combinations of 2 or 3…
Analytical
Reasoning
and
Interaction
Data
Representation
Transformation
Production,
Presentation
Dissemination
• Tech Transfer
• Report Generation
• etc
•
•
•
•
Interaction Design
Cognitive Psychology
Intelligence Analysis
etc.
Visual
Representation
Validation
and
Evaluation
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
This Talk Focuses On…
Analytical
Reasoning
and
Interaction
Data
Representation
Transformation
Production,
Presentation
Dissemination
•
•
•
•
Interaction Design
Cognitive Psychology
Intelligence Analysis
etc.
Visual
Representation
•
•
•
•
InfoVis
SciVis
Graphics
etc
Validation
and
Evaluation
• Quality Assurance
• User studies (HCI)
• etc
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Interactive Analysis + Visualization
• Most people in the visualization community believe
that interactivity is essential for visualization and visual
analytics:
– “A [visual] analysis session is more of a dialog between the
analyst and the data… the manifestation of this dialog is
the analyst’s interactions with the data representation”
[Thomas & Cook 2005]
– “Without interaction, [a visualization] technique or system
becomes a static image or autonomously animated
images” [Yi et al. 2007]
• The goal of this talk is to consider the role of
interaction in computer graphics, information
visualization, and visual analytics.
• First, we consider a stereotypical graphics application
and try adding interaction to it..
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Urban Simplification
• (left) Original model, 285k polygons
• (center) e=100, 129k polygons (45% of original)
• (right) e=1000, 53k polygons (18% of original)
R. Chang et al., Legible simplification of textured urban models. IEEE Computer Graphics and Applications, 28(3):27–36, 2008.
R. Chang et al., Hierarchical simplification of city models to maintain urban legibility. ACM SIGGRAPH 2006 Sketches, page 130 , 2006.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Urban Simplification
• Which polygons to remove?
Original Model
Our Textured Model
Simplified Model
using QSlim
Our Model
Visually different, but quantitatively similar!
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Urban Simplification
• The goal is to retain the “Image of the City”
• Based on Kevin Lynch’s concept of “Urban
Legibility” [1960]
–
–
–
–
–
Paths: highways, railroads
Edges: shorelines, boundaries
Districts: industrial, historic
Nodes: Time Square in NYC
Landmarks: Empire State building
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Urban Visualization with Semantics
• How do people think about a city?
– Describe New York…
• Response 1: “New York is large, compact, and crowded.”
• Response 2: “The area where I live there has a strong mix of
ethnicities.”
Geometric, Information, View Dependent (Cognitive)
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Urban Visualization
• Geometric
– Create a hierarchy of shapes based on the rules of legibility
• Information
– Matrix view and Parallel Coordinates show relationships between clusters and
dimensions
• View Dependence (Cognitive)
– Uses interaction to alter the position of focus
R. Chang et al., Legible cities: Focus-dependent multi-resolution visualization of urban relationships. IEEE Transactions on Visualization
and Graphics , 13(6):1169–1175, 2007
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
The Role of Interaction in Visualization
• We can use interactions to… [Yi et al. 2007]
–
–
–
–
–
–
–
Select: mark something as interesting
Explore: show me something else
Reconfigure: show me a different arrangement
Encode: show me a different representation
Abstract/Elaborate: show me more or less detail
Filter: show me something conditionally
Connect: show me related items
• In other words, we can use interactions to think.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(1) WireVis: Financial Fraud Analysis
• In collaboration with Bank of America
– Looks for suspicious wire transactions
– Currently beta-deployed at WireWatch
– Visualizes 15 million transactions over 1 year
• Uses interaction to coordinate four perspectives:
–
–
–
–
Keywords to Accounts
Keywords to Keywords
Keywords/Accounts over Time
Account similarities (search by example)
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(1) WireVis: Financial Fraud Analysis
Heatmap View
(Accounts to Keywords
Relationship)
Search by Example
(Find Similar
Accounts)
Keyword Network
(Keyword
Relationships)
Strings and Beads
(Relationships over Time)
R. Chang et al., Scalable and interactive visual analysis of financial wire transactions for fraud detection. Information Visualization,2008.
R. Chang et al., Wirevis: Visualization of categorical, time-varying data from financial transactions. IEEE VAST, 2007.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(2) Investigative GTD
• Collaboration with U. Maryland’s DHS Center of
Excellence START (Study of Terrorism And Response to
Terrorism)
– Global Terrorism Database (GTD)
– International terrorism activities from 1970-1997
– 60,000 incidents recorded over 120 dimensions
• Visualization is designed to be “investigative” in that it
is modeled after the 5 W’s:
– Who, what, where, when, and [why]
– Interaction allows the user to adjust one or more of the
W’s and see how that affects the other W’s
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(2) Investigative GTD
Who
Where
What
Evidence
Box
Original
Data
R. Chang et al., Investigative Visual Analysis of Global Terrorism, Journal of Computer Graphics Forum (Eurovis), 2008.
When
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(2) Investigative GTD:
Revealing Global Strategy
This group’s attacks
are not bounded by
geo-locations but
instead, religious
beliefs.
Its attack patterns
changed with its
developments.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(2) Investigative GTD:
Discovering Unexpected Temporal Pattern
A geographicallybounded entity in the
Philippines.
The ThemeRiver shows
its rise and fall as an
entity and its modus
operandi.
Domestic Group
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(3) iPCA: Interactive PCA
• Quick Refresher of Principle Component Analysis (PCA)
– Find most dominant eigenvectors as principle components
– Data points are re-projected into the new coordinate system
• For reducing dimensionality
• For finding clusters
• For many (especially novices), PCA is easy to understand
mathematically, but difficult to understand “semantically”.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(3) iPCA: Interactive PCA
R. Chang et al., iPCA: An Interactive System for PCA-based Visual Analytics. Computer Graphics Forum (Eurovis), 2009. To Appear.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
(3) Evaluation – iPCA vs. SAS/INSIGHT
• Results
– A bit more accurate
– People don’t “give up”
– Not faster
• Overall preference
– Using letter grades (A
through F) with “A”
representing excellent
and F a failing grade.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
If (Interactions == Thinking)…
• What is in a user’s interactions?
• If (interactions == thinking), what can we learn
from the user’s interactions?
• Is it possible to extract “thinking” from
“interactions”?
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
What is in a User’s Interactions?
Keyboard, Mouse, etc
Input
Visualization
Human
Output
Images (monitor)
• Types of Human-Visualization Interactions
– Word editing (input heavy, little output)
– Browsing, watching a movie (output heavy, little input)
– Visual Analysis (closer to 50-50)
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
What is in a User’s Interactions?
• Goal: determine if there really is “thinking” in a
user’s interactions.
Grad
Students
(Coders)
Compare!
(manually)
Analysts
Strategies
Methods
Findings
Guesses of
Analysts’
thinking
Logged
(semantic)
Interactions
WireVis
Interaction-Log Vis
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
What’s in a User’s Interactions
• From this experiment, we find that interactions contains at least:
– 60% of the (high level) strategies
– 60% of the (mid level) methods
– 79% of the (low level) findings
R. Chang et al., Recovering Reasoning Process From User Interactions. IEEE Computer Graphics and Applications, 2009.
R. Chang et al., Evaluating the Relationship Between User Interaction and Financial Visual Analysis. IEEE Symposium on VAST, 2009.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
What’s in a User’s Interactions
• Why are these so much
lower than others?
– (recovering “methods” at
about 15%)
• Only capturing a user’s
interaction in this case is
insufficient.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Lessons Learned
• We have proven that a great deal of an analyst’s “thinking” in using
a visualization is capturable and extractable.
• Using semantic interaction capturing, we might be able to collect all
the thinking of expert analysts and create a knowledge database
that is useful for
– Training: many domain specific analytics tasks are difficult to teach
– Guidance: use existing knowledge to guide future analyses
– Verification, and validation: go back and check to see if everything was
done right.
• But not all visualizations are interactive, and not all thinking is
reflected in the interactions.
– A model of how and what to capture in a visualization for extracting an
analyst’s thinking process is necessary.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Conclusion
• Interactions are important for visualization and visual
analysis
– In considering interactions, one must be aware of the
necessary speed and frame rate of the displays.
• Techniques such as simplification, LOD, or approximation can be
used.
– Interactions have been proven to help the understanding
of complex problems.
• Relevant interactions have been integrated in multiple
visualizations for different domains and demonstrated significant
impact.
– Capturing and storing analysts’ interactions have great
potential
• They can be aggregated to become a “knowledge database” that
has traditionally been difficult to create manually.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Discussion
• What interactivity is not good for:
– Presentation
– YMMV = “your mileage may vary”
• Reproducibility: Users behave differently each time.
• Evaluation is difficult due to opportunistic discoveries..
– Often sacrifices accuracy
• iPCA – SVD takes time on large datasets, use iterative
approximation algorithms such as onlineSVD.
• WireVis – Clustering of large datasets is slow. Either
pre-compute or use more trivial “binning” methods.
Intro
Vis/Gfx
Interaction
Evaluation
Discussion
• Interestingly,
– It doesn’t save you time…
– And it doesn’t make a user more
accurate in performing a task.
• However, there are empirical
evidence that using interactivity:
– Users are more engaged (don’t
give up)
– Users prefer these systems over
static (query-based) systems
– Users have a faster learning curve
• We need better measurements
to determine the “benefits of
interactivity”
Wrap-up
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Future Work
• Data Mining
• Machine
Learning
• Databases
• Information
Retrieval
• etc
Analytical
Reasoning
and
Interaction
Data
Representation
Transformation
Production,
Presentation
Dissemination
• Tech Transfer
• Report Generation
• etc
•
•
•
•
Interaction Design
Cognitive Psychology
Intelligence Analysis
etc.
Visual
Representation
•
•
•
•
InfoVis
SciVis
Graphics
etc
Validation
and
Evaluation
• Quality Assurance
• User studies (HCI)
• etc
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Future Work
• Lots of possible combinations. Are they all
meaningful?
• Of particular interest to me is “Data +
Interaction + Visualization”
– How to apply computational approaches to find
solutions that are usable by humans?
• Linear (PCA) and non-linear (manifold learning) create
dimensions that are semantically difficult to define
• Nodes within a Bayesian network are difficult to
comprehend, therefore the results difficult to take at
face value.
Intro
Vis/Gfx
Interaction
Evaluation
Wrap-up
Thank you!
[email protected]
http://www.viscenter.uncc.edu/~rchang