CS 5764 Information Visualization

1
CS 5764Information Visualization

• Dr. Chris North
• Purvi Saraiya GTA

2
Today

1. What is Information Visualization?
2. Who cares?
3. What will I learn?
4. How will I learn it?

3
1. What is Information Visualization?

• The use of computer-supported, interactive,
  visual representations of abstract data to
  amplify cognition
• Card, Mackinlay, Shneiderman

4
The Big Problem
Web, scientific datanews, products
salesshoppingcensus datasystem logsj sports
Human
Data
Data Transfer
How?
Vision 100MB/sec Aural 100KB/sec Smell Haptic
s Taste esp
5
Human Vision

• Highest bandwidth sense
• Fast, parallel
• Pattern recognition
• Pre-attentive
• Extends memory and cognitive capacity
• (Multiplication test)
• People think visually
• Brain 8 lbs, vision 3 lbs
• Impressive. Lets use it!

6
Find the Red Square
Pre-attentive
7

• Which state has highest Income? Avg?
  Distribution?
• Relationship between Income and Education?
• Outliers?

8
College Degree
Per Capita Income
9

10
Visual Representation Matters!

• Text vs. Graphics
• What if you could only see 1 states data at a
  time? (e.g. Census Bureaus website)
• What if I read the data to you?
• Graphics vs. Graphics
• depends on user tasks, data,

11
History Static Graphics
Minard, 1869
12
The Big Problem
Human
Data
Data Transfer
visualization
13
The Bigger Problem
Data
Human
Data Transfer
interactive visualization
14
Interactive Graphics

• Homefinder

15
(No Transcript)
16
(No Transcript)
17
Search Forms

• Avoid the temptation to design a form-based
  search engine
• More tasks than just search
• How do I know what to search for?
• What if theres something better that I dont
  know to search for?
• Hides the data
• Only supports QA

18
User Tasks
Excel can do this

• Easy stuff
• Min, max, average,
• These only involve 1 data item or value
• Hard stuff
• Patterns, trends, distributions, changes over
  time,
• outliers, exceptions,
• relationships, correlations, multi-way,
• combined min/max, tradeoffs,
• clusters, groups, comparisons, context,
• anomalies, data errors,
• Paths,

Visualization can do this!
19
More than just data transfer

• Glean higher level knowledge from the data
• Learn data ? knowledge

• Reveals data
• Reveals knowledge that is not necessarily
  stored in the data
• Insight!

• Hides data
• Hampers knowledge
• Nothing learned
• No insight

20
Class Motto

• Show me the data!

21
2. Who Cares?
22

• Presentation is everything

23
My Philosophy Optimization

• Computer
• Serial
• Symbolic
• Static
• Deterministic
• Exact
• Binary, 0/1
• Computation
• Programmed
• Follow instructions
• Amoral

• Human
• Parallel
• Visual
• Dynamic
• Non-deterministic
• Fuzzy
• Gestalt, whole, patterns
• Understanding
• Free will
• Creative
• Moral

• Visualization the best of both
• Impressive computation impressive cognition

24
3. What Will I Learn?

• Design interactive visualizations
• Critique existing designs and tools
• Develop visualization software
• Empirically evaluate designs
• Understand current state-of-art
• An HCI focus
• A visualization a user interface for data

25
Topics

• Information Types
• Multi-D
• 1D, 2D, 3D spatial
• Hierarchies/Trees
• Networks/Graphs
• Document collections

• Strategies
• Design Principles
• Interaction strategies
• Navigation strategies
• Visual Overviews
• Multiple Views
• Empirical Evaluation
• Development
• Theory
• High-Resolution Displays

26
GigaPixel Display
27
Related Courses

• Scientific Visualization (ESM4714)
• Computer Graphics (4204, 6xxx)
• Usability Engineering (5714)
• Research Methods (5014)
• Model Theories of HCI (5724)
• User Interface Software (5774)
• Info Storage Retrieval (5604)
• Databases (5614), Digital Libraries (6xxx)
• Data Mining (6xxx)

28
4. How will I learn it?Course Mechanics

• http//infovis.cs.vt.edu/cs5764/
• Grading (See Syllabus online)
• 60 Project
• 30 Homeworks
• 5 Paper presentation or review
• 5 Experiment class participation
• Format
• Read research papers (see web site)
• In-class discussion
• Emphasis on project

29
Research Class

• Creativity
• Open ended
• Often no right answer
• Reasoning/argument is more important
• Thinking deeply
• Self motivation, seek to excel
• Contribute to the state-of-the-art
• Jump start for thesis research, publication

30
Project

• Groups of 3 students
• Visualization for Intelligence Analysis
• Milestones
• Team choose team (due Wed!)
• Design Concept Presentation problem, lit.
  review, design, schedule (4 weeks)
• Formative Eval Initial Impl
• Final presentation final results
• Final paper publishable?

31
Project

• Groups of 3 students
• Categories
• Development design, implement, evaluate new
  visualization
• Evaluation empirical experiments with users
• Theory literature survey, synthesize theory or
  taxonomy
• Milestones
• Abstract choose team and topic (due next
  week!)
• Proposal problem, lit. review, design, schedule
• Mid-semester presentation initial results
• Final presentation final results
• Final paper publishable?

32
Paper Presentations

• 10-15 minutes
• Read paper, Present visualization
• Information type
• Visual mappings
• Show pictures / demo / video
• Strengths, weaknesses
• E.g. Scale, insight factor, user tasks

33
Presentations

• Goals
• 1 understand visualization (mappings, simple
  examples)
• 2 strengths, weaknesses
• Tips
• Time is short 10-15 min 7 slides, practice
  out loud
• Use pictures, pictures, pictures, pictures,
• Use text only to hammer key points
• The slide-sorter test
• Whats the take-home message? 2 main points
• Conclude with controversy
• Motivate!

34
Implementation detail crap

• The first step of processing requires the
  construction of several tree and graph structures
  to store the database.
• System then builds visualization of data by
  mapping data attributes of graph items to
  graphical attributes of nodes and links in the
  visualization windows on screen.
• More boring stuff nobody is ever going to read
  here or if they do they wont understand it anyway
  so why bother.
• If they do read it then they most certainly will
  not be listening to what you are saying so why
  bother give a talk? Why not just sit down and
  let everybody read your slides or just hand out
  the paper and then say thank you.
• This person needs to take Dr. Norths info vis
  class.

35
To Do

• Read CMS chapter 1 handout (pg 1-16)
• HW 1, due next Mon SequoiaView
• Form project teams
• Wed Intell Analysis exercise Projects

36
Force Adds?

• Why?
• Academic goals?
• Can you keep up?