dimensionintrovagraphicsinteractionwrap-up 1/50 data exploration, analysis, and representation:...
TRANSCRIPT
DimensionIntro VA Graphics Interaction Wrap-up1/50
Data Exploration, Analysis, and Representation: Integration through Visual Analytics
Remco Chang
UNC CharlotteCharlotte Visualization Center
DimensionIntro VA Graphics Interaction Wrap-up2/50
Problem Statement
• The growth of data is exceeding our ability to analyze them.
• The amount of digital information generated in the years 2002, 2006, 2010:– 2002: 22 EB (exabytes, 1018)– 2006: 161 EB– 2010: 988 EB (almost 1 ZB)
1: Data courtesy of Dr. Joseph Kielman, DHS2: Image courtesy of Dr. Maria Zemankova, NSF
DimensionIntro VA Graphics Interaction Wrap-up3/50
Problem Statement
• The data is often complex, ambiguous, noisy. Analysis of which requires human understanding.
– About 2 GB of digital information is being produced per person per year
– 95% of the Digital Universe’s information is unstructured
1: Data courtesy of Dr. Joseph Kielman, DHS2: Image courtesy of Dr. Maria Zemankova, NSF
DimensionIntro VA Graphics Interaction Wrap-up4/50
Example: What Does Fraud Look Like?
• Financial Institutions like Bank of America have legal responsibilities to report all suspicious activities
• Data size: approximately 200,000 transactions per day (73 million transactions per year)
• Problems:– Automated approach can only detect known patterns– Bad guys are smart: patterns are constantly changing– No single transaction appears fraudulent– Few experts: fraud detection is considered an “art”– Data is messy: lack of international standards resulting in ambiguous data
• Current methods:– 10 analysts monitoring and analyzing all transactions– Using SQL queries and spreadsheet-like interfaces– Limited to the time scale (2 weeks)
DimensionIntro VA Graphics Interaction Wrap-up5/50
Eureka: Visual Analytics!!
“Saunders, perhaps you’re getting a bit carried away with the visual analytics!”1
1: Slide courtesy of Dr. Maria Zemankova, NSF
DimensionIntro VA Graphics Interaction Wrap-up6/50
WireVis: Financial Fraud Analysis
• In collaboration with Bank of America– Looks for suspicious wire transactions– Currently beta-deployed at WireWatch– Visualizes 7 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)
DimensionIntro VA Graphics Interaction Wrap-up7/50
WireVis: Financial Fraud Analysis
Heatmap View(Accounts to Keywords Relationship)
Strings and Beads(Relationships over Time)
Search by Example (Find Similar Accounts)
Keyword Network(Keyword Relationships)
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.
DimensionIntro VA Graphics Interaction Wrap-up8/50
What is 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.
DimensionIntro VA Graphics Interaction Wrap-up9/50
Individually Not Unique
Analytical Reasoning
and Interaction
Visual Representation
Production, Presentation
Dissemination
Data Representation Transformation
Validation and Evaluation
• Data Mining• Machine
Learning• Databases• Information
Retrieval• etc
• Tech Transfer• Report Generation• etc
• Quality Assurance• User studies (HCI)• etc
• Interaction Design• Cognitive Psychology• Intelligence Analysis• etc.
• InfoVis• SciVis• Graphics• etc
DimensionIntro VA Graphics Interaction Wrap-up10/50
In Combinations of 2 or 3…
Analytical Reasoning
and Interaction
Visual Representation
Production, Presentation
Dissemination
Data Representation Transformation
Validation and Evaluation
• Data Mining• Machine
Learning• Databases• Information
Retrieval• etc
• InfoVis• SciVis• Graphics• etc
DimensionIntro VA Graphics Interaction Wrap-up11/50
In Combinations of 2 or 3…
Analytical Reasoning
and Interaction
Visual Representation
Production, Presentation
Dissemination
Data Representation Transformation
Validation and Evaluation
• Interaction Design• Cognitive Psychology• Intelligence Analysis• etc.
• Tech Transfer• Report Generation• etc
DimensionIntro VA Graphics Interaction Wrap-up12/50
Case Study on WireVis
Analytical Reasoning
and Interaction
Visual Representa
tion
Production, Presentatio
n Disseminati
on
Data Representat
ion Transformat
ion
Validation and
Evaluation
• User Centric– Designed system based on domain
expertise• Visual Interface
– Multiple coordinated views that link multiple dimensions
• Interactive– Overview, drill-down, reclustering
• Data Clustering– Clustering by accounts, and search
by example• Production
– Connected to a live database and beta-deployed at BofA
• (Validation)– Expert evaluation
DimensionIntro VA Graphics Interaction Wrap-up13/50
This Talk Focuses On…
Analytical Reasoning
and Interaction
Visual Representation
Production, Presentation
Dissemination
Data Representation Transformation
Validation and Evaluation
• Interaction Design• Cognitive Psychology• Intelligence Analysis• etc.
• InfoVis• SciVis• Graphics• etc
• Data Mining• Machine
Learning• Databases• Information
Retrieval• etc
DimensionIntro VA Graphics Interaction Wrap-up14/50
Visual Analytics, A Graphics Perspective
DimensionIntro VA Graphics Interaction Wrap-up15/50
Visual Analytics, A Graphics Perspective
• Master’s Thesis -- – Simulating dynamic motion
based on kinematic motion• Jiggling of muscles
– Skinnable Mesh • Volumetric deformation
– Compared 3 types of mass-spring systems• Regular (unconstrained) mass-
spring• Reduced degree of freedom• Approximate finite element
method with implicit integration
• Is this applicable beyond graphics and simulation?
R. Chang, Simulation Techniques for Deformable Animated Characters. Master’s Thesis, Brown University, 2000.
DimensionIntro VA Graphics Interaction Wrap-up16/50
From Graphics to Visual Analytics:An Example in 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.
DimensionIntro VA Graphics Interaction Wrap-up17/50
Urban Simplification
• Which polygons to remove?
Original Model Simplified Model using QSlim
Our Textured Model Our Model
Visually different, but quantitatively similar!
DimensionIntro VA Graphics Interaction Wrap-up18/50
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
DimensionIntro VA Graphics Interaction Wrap-up19/50
Algorithm to Preserve Legibility
• Identify and preserve Paths and Edges• Create logical Districts and Nodes• Simplify model while preserving Paths, Edges, Districts, and
Nodes• Hierarchically apply appropriate amount of texture • Highlight Landmarks and choose models to render
DimensionIntro VA Graphics Interaction Wrap-up20/50
Algorithm for Preserving Legibility• Paths & Edges
– Hierarchical (single-link) clustering
• Nodes– Merging clusters– Polyline simplification
using convex hulls
• Landmarks– Pixel-based skyline
preservation
• That’s pretty good, right?
DimensionIntro VA Graphics Interaction Wrap-up21/50
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)
DimensionIntro VA Graphics Interaction Wrap-up22/50
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
DimensionIntro VA Graphics Interaction Wrap-up23/50
Probe-based Interactions
• Using Probes allows for comparing multiple regions-of-interest simultaneously
R. Chang et al., Multi-focused geospatial analysis using probes. Visualization and Computer Graphics, IEEE Transactions on, 14(6):1165–1172, Nov.-Dec. 2008.
DimensionIntro VA Graphics Interaction Wrap-up24/50
Urban VisualizationGraphics + Visual Analytics
• Applying graphics approaches– Data transformation– Screen-based metrics– Hardware acceleration
• Applying visual analytics principles– Multi-dimensional data
representation– Interactive exploration– Broader applicability
DimensionIntro VA Graphics Interaction Wrap-up25/50
Extending Visual Analytics Principles
• Global Terrorism Database– Application of the
investigative 5 W’s
• Bridge Maintenance – Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methods
Where
When
Who
What
Original Data
EvidenceBox
R. Chang et al., Investigative Visual Analysis of Global Terrorism, Journal of Computer Graphics Forum, 2008.
DimensionIntro VA Graphics Interaction Wrap-up26/50
Extending Visual Analytics Principles
• Global Terrorism Database– Application of the
investigative 5 W’s
• Bridge Maintenance – Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methodsR. Chang et al., An Interactive Visual Analytics System for Bridge Management, Journal of Computer Graphics Forum, 2010. To Appear.
DimensionIntro VA Graphics Interaction Wrap-up27/50
Extending Visual Analytics Principles
• Global Terrorism Database– Application of the
investigative 5 W’s
• Bridge Maintenance – Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methodsR. Chang et al., Interactive Coordinated Multiple-View Visualization of Biomechanical Motion Data , IEEE Vis (TVCG) 2009.
DimensionIntro VA Graphics Interaction Wrap-up28/50
Human + ComputerA Mixed-Initiative Perspective• So far, our approach is mostly user-driven
• Human vs. Artificial IntelligenceGarry Kasparov vs. Deep Blue (1997)– Computer takes a “brute force” approach without analysis– “As for how many moves ahead a grandmaster sees,” Kasparov concludes: “Just one,
the best one”
• Artificial Intelligence vs. Augmented IntelligenceHydra vs. Cyborgs (1998)– Grandmaster + 1 computer > Hydra (equiv. of Deep Blue)– Amateur + 3 computers > Grandmaster + 1 computer1
• How to systematically repeat the success? – Unsupervised machine learning + User– User’s interactions with the computer
1. http://www.collisiondetection.net/mt/archives/2010/02/why_cyborgs_are.php
Computer Translation Human
DimensionIntro VA Graphics Interaction Wrap-up29/50
Human + Computer:Dimension Reduction – Lost in Translation• Biomechanical motion analysis revisited…
– 6 degrees of freedom (x, y, z rotation and x, y, z translation)– One single joint
• Applying a non-linear dimension reduction method (manifold learning)– Isomap– MDS embedding
• We found:– 3 latent dimensions– 2 of which are ambiguous…
DimensionIntro VA Graphics Interaction Wrap-up30/50
Human + Computer:Dimension Reduction• Non-linear is too hard. Let’s start with a classical linear approach,
principle component analysis (PCA).
• Quick Refresher of PCA– Find most dominant eigenvectors as principle components– Data points are re-projected into the new coordinate system
• For many (especially novices), PCA is easy to understand mathematically, but difficult to understand “semantically”.
DimensionIntro VA Graphics Interaction Wrap-up31/50
Human + Computer:Exploring Dimension Reduction: iPCA
R. Chang et al., iPCA: An Interactive System for PCA-based Visual Analytics. Computer Graphics Forum (Eurovis), 2009.
DimensionIntro VA Graphics Interaction Wrap-up32/50
Human + Computer: Comparing iPCA to SAS/INSIGHT
• Results– A bit more accurate– Not faster– People don’t “give up”– Users seem to understand the
intuition behind PCA better
• Overall preference– Using letter grades (A through
F) with “A” representing excellent and F a failing grade.
• A lot more work needs to be done…
DimensionIntro VA Graphics Interaction Wrap-up33/50
Human + Computer:User Interactions• 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.
DimensionIntro VA Graphics Interaction Wrap-up34/50
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”?
DimensionIntro VA Graphics Interaction Wrap-up35/50
What is in a User’s Interactions?
• Goal: determine if there really is “thinking” in a user’s interactions.
Analysts
GradStudents(Coders)
Logged(semantic) Interactions
Compare!(manually)
StrategiesMethodsFindings
Guesses ofAnalysts’ thinking
WireVis Interaction-Log Vis
DimensionIntro VA Graphics Interaction Wrap-up36/50
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.
DimensionIntro VA Graphics Interaction Wrap-up37/50
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.
DimensionIntro VA Graphics Interaction Wrap-up38/50
User Interactions, An Analytical Approach• Now that we’ve shown that (interaction ~= thinking)
– Can we automate the process?
• Formulate every user interaction as a fixed-length vector (Design Galaries [Marks et al. Siggraph 97]). For example,
• User interaction in the left application can be represented as a single dimensional vector <P>• User interaction in the right application can be represented as a two dimensional vector <P,
S>
DimensionIntro VA Graphics Interaction Wrap-up39/50
Human + Computer:User Interactions – Lessons Learned• We have proven that a great deal of an analyst’s “thinking” in using a
visualization is capturable and extractable.– Although the study is limited in scope, it establishes a foundation for interaction-
capturing related research
• With 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: 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 process is necessary.
• Automating the process of extracting thinking is the key.– By formulating user interactions as high dimensional vectors, we can apply analytical
methods– The assumption is that the vectors are fixed-length. Is this always true?
DimensionIntro VA Graphics Interaction Wrap-up40/50
Conclusion
Analytical Reasoning
and Interaction
Visual Representa
tion
Production, Presentatio
n Disseminati
on
Data Representat
ion Transformat
ion
Validation and
Evaluation
• Visual Analytics is a growing new area that is looking to address some pressing needs– Too much (messy) data, too little
time
• By combining strengths and findings in existing disciplines, we have demonstrated that– There are some great benefits– But there are also some difficult
challenges
DimensionIntro VA Graphics Interaction Wrap-up41/50
Summary of Contribution• Contributions
– Visual Representation• Urban modeling and visualization
– Interaction + Visual Representation• Role of interactivity in visual thinking• Applying principles to real-world problems
such as financial analytics, terrorism studies, bridge management, biomechanical motion analysis, etc.
– Interaction + Data Analysis• Exploring principle component analysis
– Evaluation + Interaction• Proposed a new learning-based evaluation
methods
• In particular, my background in computer graphics helps the development of a human + computer research agenda
DimensionIntro VA Graphics Interaction Wrap-up42/50
Future Work (Funded Projects)• NSF SciSIP:
– Title: A Visual Analytics Approach to Science and Innovation Policy. • PI: William Ribarsky, Co-PIs: Jim Thomas, Remco Chang, Jing Yang. • $746,567. 2009-2012 (3 years).
– Abstract: developing metrics and visual tools for identifying patterns in science policies.
• NSF/DOD (Minerva Initiative): – Title: Collaborative Project: Terror, Conflict Processes, Organizations, & Ideologies: Completing the
Picture. • PI: Remco Chang• $100,000. 2009-2010 (2 years).
– Abstract: design and develop visual analytical tools to identifying the causal relationships in government policies and domestic conflicts.
• DHS International Program: – Title: Deriving and Applying Cognitive Principles for Human/Computer Approaches to Complex Analytical
Problems. • PI: William Ribarsky, Co-PIs: Brian Fisher, Remco Chang, John Dill. • $200,000. 2009-2010 (1 year).
– Abstract: identifying new evaluation methods for visual analytical systems, and applying computational methods for analyzing user interactions.
• Quantitative Analysis Division at Bank of America (Dr. Agus Sudjianto)– Exploration and analysis of financial risk
DimensionIntro VA Graphics Interaction Wrap-up43/50
Future Work (On-going Collaborations)• With NSF FODAVA Center at Georgia Tech (Dr. Haesun Park, director)
– Interpreting user interactions to affecting machine learning algorithms– Visual PCA: using perceptual metrics to finding principle components– Applying perceptual constraint to dimension reduction: for animating temporal data
in dimension reduction, find methods to maintain hysteresis
• With University of Kentucky (Drs. Judy Goldsmith, Jinze Liu, Phillip Chang, MD)– Integrating data mining (KDD), POMDP, and visual analytics to preventing sepsis by
identifying biomarkers (Proposal in submission to NSF CDI)
• With geographer and architect at UNC Charlotte (Dr. Jean-Claude Thill and Eric Sauda)– Designing computational methods for identifying neighborhood characteristics
(Proposal in submission to NSF IIS)– Applying the UrbanVis system to analyzing crime (proposal in preparation for DOJ/NIJ)
• With Virginia Tech (Dr. Chris North)– Developing a research agenda for analytic provenance
DimensionIntro VA Graphics Interaction Wrap-up44/50
Journal Publications (16)• Urban Visualization
– R. Chang, T. Butkiewicz, C. Ziemkiewicz, Z. Wartell, N. Pollard, and W. Ribarsky. Legible simplification of textured urban models. IEEE Computer Graphics and Applications, 28(3):27–36, 2008.
– T. Butkiewicz, R. Chang, Z. Wartell, and W. Ribarsky. Visual analysis of urban change. Computer Graphics Forum, 27(3):903–910, 2008.– T. Butkiewicz, R. Chang, W. Ribarsky, and Z. Wartell. Understanding Dynamics of Geographic Domains, chapter Visual Analysis of Urban Terrain Dynamics,
pages 151– 169. CRC Press/Taylor and Francis, 2007.– R. Chang, G. Wessel, R. Kosara, E. Sauda, and W. Ribarsky. Legible cities: Focus-dependent multi-resolution visualization of urban relationships. Visualization
and Computer Graphics, IEEE Transactions on, 13(6):1169–1175, Nov.-Dec. 2007.• Visualization and Visual Analytics
– X. Wang, W. Dou, S.E. Chen, W. Ribarsky, and R. Chang. An interactive visual analytics system for bridge management. Computer Graphics Forum (Eurovis 2010), 2010. Conditional acceptance.
– D. Keefe, M. Ewert, W. Ribarsky, and R. Chang. Interactive coordinated multiple-view visualization of biomechanical motion data. Visualization and Computer Graphics, IEEE Transactions on (IEEE Visualization Conference), 15(6):1383–1390, 2009
– X. Wang, D.H. Jeong, W. Dou, S.W. Lee, W. Ribarsky, and R. Chang. Defining and applying knowledge conversion processes to a visual analytics system. Computers & Graphics, July 2009. [Online] doi:10.1016/j.cag.2009.06.004
– D.H. Jeong, C. Ziemkiewicz, B. Fisher, W. Ribarsky, and R. Chang. iPCA: An interactive system for PCA-based visual analytics. Computer Graphics Forum, 28(3):767–774, 2009.
– R. Chang, C. Ziemkiewicz, T.M. Green, and W. Ribarsky. Defining insight for visual analytics. IEEE Computer Graphics and Applications, 29(2):14–17, 2009.– R. Chang, A. Lee, M. Ghoniem, R. Kosara, W. Ribarsky, J. Yang, E. Suma, C. Ziemkiewicz, D. Kern, and A. Sudjianto. Scalable and interactive visual analysis of
financial wire transactions for fraud detection. Information Visualization, 7:63–76(14), 2008.– X. Wang, E. Miller, K. Smarick, W. Ribarsky, and R. Chang. Investigative visual analysis of global terrorism database. Computer Graphics Forum, 27(3):919–
926, 2008.• Interaction & Provenance
– W. Pike, J. Stasko, R. Chang, and T. O’Connell. Science of interaction. Information Visualization, 8:263–274, 2009.– W. Dou, D.H. Jeong, F. Stukes, W. Ribarsky, H. Lipford, and R. Chang. Recovering reasoning process from user interactions. IEEE Computer Graphics and
Applications, 29(3):52–61, 2009• VR & Interface Designs
– T. Butkiewicz, R. Chang, Z. Wartell, and W. Ribarsky. Alleviating the modifiable areal unit problem with probe-based geospatial analyses. Computer Graphics Forum (Eurovis 2010), 2010. Conditional acceptance
– T. Butkiewicz, W. Dou, Z. Wartell, W. Ribarsky, and R. Chang. Multi-focused geospatial analysis using probes. Visualization and Computer Graphics, IEEE Transactions on, 14(6):1165–1172, Nov.-Dec. 2008.
– D.H. Jeong, C. Song, R. Chang, and L. Hodges. User experimentation: An evaluation of velocity control techniques in immersive virtual environments. Springer-Verlag Virtual Reality, 13(1):41–50, Mar. 2009.
DimensionIntro VA Graphics Interaction Wrap-up45/50
Conference/Workshop (22)• R. Chang, C. Ziemkiewicz, R. Pyzh, J. Kielman, and W. Ribarsky. Learning-based evaluation of visual analytics systems. In ACM SIGCHI BELIV Workshop, 2010. Conditional
acceptance.• D. H. Jeong, T. Green, W. Ribarsky, and R. Chang. Comparative evaluation of two interface tools in performing visual analytics tasks. In ACM SIGCHI BELIV Workshop, 2010.
Conditional acceptance.• G. Wessel, E. Unruh, R. Chang, and E. Sauda. Urban user interface: Urban legibility reconsidered. In Southwest ACSA, 2010.• D. H. Jeong, W. Dou, W. Ribarsky, and R. Chang. Knowledge-oriented refactoring in visualization. In IEEE Visualization Workshop on Refactoring Visualization From Experience,
2009.• D. H. Jeong, W. Ribarsky, and R. Chang. Designing a PCA-based collaborative visual analytics system. In IEEE Visualization Workshop on Collaborative Visualization, 2009.• W. Dou, D. H. Jeong, F. Stukes, W. Ribarsky, H. Lipford, and R. Chang. Comparing usage patterns of domain experts and novices in visual analytical tasks. In ACM SIGCHI
Sensemaking Workshop 2009.• X. Wang, W. Dou, R. Vatcha, W. Liu, S. E. Chen, S. W. Lee, R. Chang, and W. Ribarsky. Knowledge integrated visual analysis of bridge safety and maintenance. In SPIE 2009.• X. Wang, W. Dou, W. Ribarsky, and R. Chang. Integration of heterogeneous processes through visual analytics. In SPIE 2009,.• M. Butkiewicz, T. Butkiewicz, W. Ribarsky, and R. Chang. Integrating timeseries visualizations within parallel coordinates for exploratory analysis of incident databases. SPIE
2009.• T. Butkiewicz, D. H. Jeong, W. Ribarsky, and R. Chang. Hierarchical multitouch selection techniques for collaborative geospatial analysis. In SPIE Defense, Security and Sensing
2009.• D. H. Jeong, R. Chang, and W. Ribarsky. An alternative definition and model for knowledge visualization. In IEEE Visualization Workshop on Knowledge Assisted Visualization,
2008.• X. Wang, W. Dou, S. W. Lee, W. Ribarsky, and R. Chang. Integrating visual analysis with ontological knowledge structure. In IEEE Workshop on Knowledge Assisted Visualization,
2008.• D. H. Jeong, W. Dou, F. Stukes, W. Ribarsky, H. Lipford, and R. Chang. Evaluating the relationship between user interaction and financial visual analysis. In Visual Analytics
Science and Technology. IEEE Symposium on, 2008.• G. Wessel, R. Chang, and E. Sauda. Towards a new (mapping of the) city: Interactive, data rich modes of urban legibility. In Association for Computer Aided Design in
Architecture, 2008.• G. Wessel, R. Chang, and E. Sauda. Visualizing GIS: Urban form and data structure. Seeking the City: Visionaries on the Margins, ACSA, 2008.• G. Wessel, E. Sauda, and R. Chang. Urban visualization: Urban design and computer visualization. In CAADRIA 2008.• T. Butkiewicz, R. Chang, Z. Wartell, and W. Ribarsky. Visual analysis for live lidar battlefield change detection. SPIE, 2008.• J. Jones, R. Chang, T. Butkiewicz, and W. Ribarsky. Visualizing uncertainty for geographical information in the global terrorism database. SPIE, 2008.• A. Godwin, R. Chang, R. Kosara, and W. Ribarsky. Visual analysis of entity relationships in the global terrorism database. SPIE, 2008.• T. Butkiewicz, R. Chang, Z. Wartell, and W. Ribarsky. Analyzing sampled terrain volumetrically with regard to error and geologic variation. SPIE, 2007.• R. Chang, M. Ghoniem, R. Kosara, W. Ribarsky, J. Yang, E. Suma, C. Ziemkiewicz, D. Kern, and A. Sudjianto. Wirevis: Visualization of categorical, time-varying data from financial
transactions. In Visual Analytics Science and Technology, 2007, IEEE Symposium on, 2007.• R. Chang, T. Butkiewicz, C. Ziemkiewicz, Z. Wartell, N. Pollard, and W. Ribarsky. Hierarchical simplification of city models to maintain urban legibility. In SIGGRAPH ’06: ACM
SIGGRAPH 2006 Sketches, 2006.• R. Chang, R. Kosara, A. Godwin, and W. Ribarsky. Towards a role of visualization in social modeling. AAAI 2009 Spring Symposium on Technosocial Predictive Analytics, 2009.• G. Wessel, E. Sauda, and R. Chang. Mapping understanding:Transforming topographic maps into cognitive maps. GeoVis Hamburg Workshop, 2009.
DimensionIntro VA Graphics Interaction Wrap-up46/50
Acknowledgement
Bill Ribarsky Zach Wartell
Dong Hyun Jeong, Tom Butkiewicz, Xiaoyu Wang, Wenwen Dou, Tera Green
From the Data Visualization Group (DVG) at UNC Charlotte
DimensionIntro VA Graphics Interaction Wrap-up47/50
Acknowledgement
Eric Sauda Jean-Claude Thill
From the Urban Visualization Group at UNC Charlotte
Ginette Wessel Elizabeth Unruh
DimensionIntro VA Graphics Interaction Wrap-up48/50
Acknowledgement
More Collaborators…
Nancy Pollard, Evan Suma, Heather Lipford, Dan Keefe, Caroline Ziemkiewicz, Robert Kosara, Mohammad Ghoniem Clockwise, starting on the left:
DimensionIntro VA Graphics Interaction Wrap-up49/50
Acknowledgement
• And many many others…
Joseph Kielman, Bill Pike, Theresa O'Connell, Seok-Won Lee, Brian Fisher, Alvin Lee, Jing Yang, Daniel Kern, Agust Sudjianto, Erin Miller, Kathleen Smarick, Felesia Stukes, Marcus Ewert, Larry Hodges, Michael Butkiewicz, Josh Jones, Alex Godwin, Edd Hauser, Shenen Chen, Bill Tolone, Wanqiu Liu, Rashna Vatcha
DimensionIntro VA Graphics Interaction Wrap-up50/50
Thank you!
[email protected]://www.viscenter.uncc.edu/~rchang
DimensionIntro VA Graphics Interaction Wrap-up51/50
Backup Slides – Professional Activities
DimensionIntro VA Graphics Interaction Wrap-up52/50
Grants Awarded (3)
• NSF SciSIP: – A Visual Analytics Approach to Science and Innovation Policy.
• PI: William Ribarsky, Co-PIs: Jim Thomas, Remco Chang, Jing Yang. • $746,567. 2009-2012 (3 years).
• NSF BCS: – Collaborative Project: Terror, Conflict Processes, Organizations, & Ideologies: Completing the Picture.
• PI: Remco Chang• $100,000. 2009-2010 (2 years).
• DHS International Program: – Deriving and Applying Cognitive Principles for Human/Computer Approaches to Complex Analytical
Problems. • PI: William Ribarsky, Co-PIs: Brian Fisher, Remco Chang, John Dill. • $200,000. 2009-2010 (1 year).
• In Submission / Preparation– 2 other NSF proposals are pending reviews– 1 DOH/NIJ preparation
DimensionIntro VA Graphics Interaction Wrap-up53/50
Professional Activities• Committee / Panelists
– Program Committee: IEEE Conference on Visual Analytics, 2010– Program Committee: SIG CHI Workshop on BELIV, 2010– Program Committee: AAAI Spring-09 Symposium on Technosocial Predictive Analytics, 2009– Panelist: 3rd Annual DHS University Summit. Panel: Research to Reality, 2009– Panelist: 3rd Annual DHS University Summit. Panel: Visual Analytics and Discrete Science Integration into the DHS Center of Excellence
Program, 2009• Invited Talks
– Dec 13, 2006 Google Inc. Simplification of Urban Models based on Urban Legibility– July 6, 2007 Naval Research Lab. Urban Visualization– Oct 4, 2007 Charlotte Viscenter. Urban Visualization– Oct 17, 2007 Charlotte Metropolitan GIS Users Group. GIS and Urban Visualization– Nov 19, 2007 START Center at University of Maryland. Integrated Visual Analysis of the Global Terrorism Database– Nov 29, 2007 Charlotte Viscenter. Integrated Visual Analysis of the Global Terrorism Database– Jan 25, 2008 DoD/DHS Social Science Modeling and Information Visualization Symposium. Social Science and Information Visualization on
Terrorism and Multimedia– May 14, 2008 Charlotte Metropolitan GIS User Group. Multi-Focused Geospatial Analysis Using Probes– Aug 27, 2008 DoD/DHS Symposium for Overcoming the Information Challenge in Federated Analysis: From Concept to Practice. Roadmap of
Visualization– Mar 19, 2009 DHS University Summit. Panel: Research to Reality– Mar 19, 2009 DHS University Summit. Panel: Visual Analytics and Discrete Science Integration into the DHS Center of Excellence Program– Apr 27, 2009 University of Kentucky. Thinking Interactively with Visualization– May 29, 2009 University of Victoria. Thinking Interactively with Visualization– Jul 28, 2009 Pacific Northwest National Lab. Thinking Interactively with Visualization– Jul 30, 2009 Microsoft Research. Thinking Interactively with Visualization– Aug 19, 2009 National Visual Analytic Consortium. What Are Your Interactions Doing For Your Visualization?– Sep 30, 2009 University of Kentucky (Grand Rounds at the Department of Surgery). Preventing– Sepsis: Artificial Intelligence, Knowledge Discovery, and Visualization– Jan 21, 2010 Charlotte Viscenter. UrbanVis Research Group: Urban Analytics– Feb 25, 2010 University of Georgia (AI Institute). Thinking Interactively with Visualization
DimensionIntro VA Graphics Interaction Wrap-up54/50
Backup Slides – Urban Simplification
DimensionIntro VA Graphics Interaction Wrap-up55/50
Identifying and Preserving Paths and Edges
DimensionIntro VA Graphics Interaction Wrap-up56/50
• Single-Link Clustering – Iteratively groups the “closest” clusters
together based on Euclidean distance– produces a binary tree (dendrogram)– Penalizes large clusters to create a more
balanced tree
Identifying and PreservingPaths and Edges (1)
a b c d e f
bc de
def
bcdef
abcdef
abc
DimensionIntro VA Graphics Interaction Wrap-up57/50
Identifying and PreservingPaths and Edges (2)
DimensionIntro VA Graphics Interaction Wrap-up58/50
Creating logical Districts and Nodes
DimensionIntro VA Graphics Interaction Wrap-up59/50
Creating logical Districts and Nodes (1)
DimensionIntro VA Graphics Interaction Wrap-up60/50
• Merge two clusters by combining footprints
Creating logical Districts and Nodes (2)
• (c) The resulting “Merged Hull”• (d) The Introduced Error, or “Negative Space”
(a) (b) (c) (d)
DimensionIntro VA Graphics Interaction Wrap-up61/50
Simplification while preserving Paths, Edges, Nodes, and Districts
DimensionIntro VA Graphics Interaction Wrap-up62/50
Simplification while preserving Paths, Edges, Nodes, and Districts (1)
6000 edges 1000 edges
Demo!
DimensionIntro VA Graphics Interaction Wrap-up63/50
Simplification while preserving Paths, Edges, Nodes, and Districts (2)
• After the polylines have been simplified– Create “Cluster Meshes”
– The height of the Cluster Mesh is the median height of all buildings in the cluster
DimensionIntro VA Graphics Interaction Wrap-up64/50
Hierarchical Textures
DimensionIntro VA Graphics Interaction Wrap-up65/50
• Each Cluster Mesh contains 6 textures– 1 Side Texture
– 1 top-down view of the roof texture
– 4 roof textures from 4 angles (south, west, east, north)
Hierarchical Textures (1)
Side texture
Top-down South West East North
DimensionIntro VA Graphics Interaction Wrap-up66/50
• Clusters are divided into “bins” based on their visual importance
• Each bin contains a texture atlas
• Texture atlases from all bins have the same dimension
Hierarchical Textures (2)
n/2 n/4 n/8
….
…
DimensionIntro VA Graphics Interaction Wrap-up67/50
Runtime Levels of Detail
DimensionIntro VA Graphics Interaction Wrap-up68/50
• Starting with the root node of the dendrogram– Approximate the “Negative
Space” as a 3D box – shown as the red box
– Project the visible sides of the box onto screen space
– Reject if the number of pixel is above a user-defined tolerance
Runtime Levels of Detail
a b c d e f
bc de
def
abcdef
abc
DimensionIntro VA Graphics Interaction Wrap-up69/50
Landmark and Skyline Preservation (1)
Original Skyline With LandmarkPreservation
Without LandmarkPreservation
DimensionIntro VA Graphics Interaction Wrap-up70/50
– Project a user-defined pixel tolerance (α) onto the top of each cluster
– If any building within that cluster is taller than the projected tolerance (shown in green), it is drawn separately from the cluster mesh.
Landmark and Skyline Preservation (2)
DimensionIntro VA Graphics Interaction Wrap-up71/50
Results
DimensionIntro VA Graphics Interaction Wrap-up72/50
(2) Investigative GTD
Where
When
Who
What
Original Data
EvidenceBox
R. Chang et al., Investigative Visual Analysis of Global Terrorism, Journal of Computer Graphics Forum (Eurovis), 2008.
DimensionIntro VA Graphics Interaction Wrap-up73/50
WHY?
This group’s attacks are not bounded by geo-locations but instead, religious beliefs.
Its attack patterns changed with its developments.
(2) Investigative GTD: Revealing Global Strategy
DimensionIntro VA Graphics Interaction Wrap-up74/50
Domestic Group
A geographically-bounded entity in the Philippines.
The ThemeRiver shows its rise and fall as an entity and its modus operandi.
(2) Investigative GTD:Discovering Unexpected Temporal Pattern
DimensionIntro VA Graphics Interaction Wrap-up75/50
What is in a User’s Interactions?
• 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)
Visualization HumanOutput
Input
Keyboard, Mouse, etc
Images (monitor)
DimensionIntro VA Graphics Interaction Wrap-up76/50
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.
DimensionIntro VA Graphics Interaction Wrap-up77/50
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”