Bongshin Lee
Researcher
Visualization and Interaction Research Group
Microsoft Research

Bongshin Lee
Computer Science Department
Human-Computer Interaction Lab
University of Maryland

Biodiversity databases have become widely available to the public and to other researchers
Trees and graphs are commonly used information structures
Animal classification, food webs, gene ontology, …
Visual representations exploit human visual processing to reduce the cognitive load
Better interactive tools are needed for visual data exploration

PaperLens
TaxonTree
NSF-funded project at UMD with
- Cyndy Parr
- Dana Campbell
- Ben Bederson

Biodiversity is the extraordinary variety of all life on Earth - from genes to species to entire ecosystems.
-- Smithsonian Institution Monitoring and Assessment of Biodiversity Program

Complex databases
Organism names
Habitats
Interactions
Conservation status
Reproductive parameters
…
Non-expert users
Policy makers
Land-use planners
Educators
Students
Laypersons
…

- Visualizing the Taxonomic Hierarchy http://www.cs.umd.edu/hcil/biodiversity
Hierarchy of 200,000 animal names
Integration of public/private sources
Extension of SpaceTree
[Plaisant et al.
, InfoVis ’02]
Interactive tree visualization
Integrated searching and browsing
Animation
B. Lee, C.S. Parr, D. Campbell, B.B. Bederson (2004)
How Users Interact with Biodiversity Information Using TaxonTree
Proceedings of AVI 2004, pp. 320-327
C.S. Parr, B. Lee, D. Campbell, B.B. Bederson (2004)
Visualizations for Taxonomic and Phylogenetic Trees
Bioinformatics, Vol. 20, No. 17, pp. 2997-3004

Collaborative design
Target audience
Undergraduates in biology class
Transitioning from novice to expert
Undergraduate design partners
Domain-specific visualization
Scaling up

Demo
Written in Java with Piccolo.Java

Qualitative study with 18 biology students
Goals
Characterize users of biodiversity domain
Investigate usability and interaction preferences
Examine information understanding
Results
Interaction with the tree is very intuitive
Browse rather than search
Make inferences using tree structure
Interested users want more control over interaction

Deployed for the public
University of Michigan’s Animal Diversity Web (ADW), http://animaldiversity.ummz.umich.edu
60,000 visits/day
Biologists love TaxonTree
Phylogeny of Lepidoptera Project (LepTree), http://www.leptree.net
California Academy of Sciences’ AntWeb, http://www.antweb.org
Cyberinfrastructure for Phylogenetic Research
(CIPRES), http://www.phylo.org
Science Environment for Ecological Knowledge
(SEEK), http://seek.ecoinformatics.org

Interactive tree visualization can be applied to the biodiversity domain for a broad audience
Integrated searching and browsing with animation
Incremental Exploration

Intern project at MSR with
- Mary Czerwinski, MSR
- George Robertson, MSR
- Ben Bederson, UMD
PaperLens
TaxonTree

InfoVis 2004 Contest
The history of InfoVis
155 papers by 315 authors for 8 years (1995~2002)
Tasks
Characterize the research areas and their evolution
Where does a particular author fit within the research areas?
What are the relationships between two or more or all authors?
http:// www.cs.umd.edu/hcil/iv04contest

- Non Node-link Graph Visualization http://www.cs.umd.edu/hcil/paperlens
Multiple small and simple components
Abstract overview of data
Show relationships dynamically
Winner of InfoVis 2004 Contest
Two design iterations: InfoVis and CHI
B. Lee, M. Czerwinski, G. Robertson, B.B. Bederson (2005)
Understanding Research Trends in Conferences using PaperLens
Extended Abstracts of CHI 2005, pp. 1069-1072
B. Lee, M. Czerwinski, G. Robertson, B.B. Bederson (2004)
Understanding Eight Years of InfoVis Conferences using PaperLens
Posters Compendium of InfoVis 2004, pp. 53-54
Intern Project at Microsoft Research

• 23 years (1982-2004)
• 4073 papers
• 6300 authors
• 15 topics
Written in C# with Piccolo.NET
Demo

Qualitative study with the InfoVis dataset
8 researchers (including 1 pilot subject)
16 tasks
Results
Most tasks were performed in less than 20 seconds
Usability Issues
Author search
Consistency
Screen usage

-- Visualizing food webs
Project at UMD with
- Cyndy Parr
- Ben Bederson

NetLens: Extension of PaperLens
General: “Content-Actor” data model
Scalable: Relational database, simple components
Implemented by Hyunmo Kang (HCIL, UMD)
H. Kang, C. Plaisant, B. Lee, B.B. Bederson (2006)
NetLens: Iterative Exploration of Content-Actor Network Data
To appear in Proceedings of VAST 2006, invited to Information Visualization Journal
H. Kang, C. Plaisant, B. Lee, B.B. Bederson (2006)
Exploring Content-Actor Paired Network Data using Iterative Query Refinement with NetLens
Proceedings of JCDL 2006, pp. 372 (Demo)
Node-link graph visualization has its place
Integrate it with EcoLens/NetLens

For more information
• http://www.cs.umd.edu/hcil/biodiversity
• http://www.cs.umd.edu/hcil/graphvis
Interactive visualization helps data exploration and understanding
TaxonTree poses little difficulty even for novice users can be applied to other taxonomies
PaperLens novel alternative to common node-link graph visualizations its concept can be applied to biodiversity domain

© 2006 Microsoft Corporation. All rights reserved.
Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation.
Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation.
MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.