30
Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of Md May 31, 2001

Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

  • View
    215

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

Musings on Treemaps

John StaskoInformation Interfaces Research GroupCollege of Computing / GVU CenterGeorgia Institute of Technology

HCIL, Univ. of Md.May 31, 2001

Page 2: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

2

Background

Early UNIX implementations of Treemap

XTangoanimationusingtechnique

Current UNIX tool

Page 3: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

3

Treemap Variation

What if we used a radial expansion (with root at center) rather than a rectangular expansion?

Add explicit representation of parent/directory

Page 4: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

4

Sunburst

Visualizing fileand directorystructures

Root dir at centerColor - file typeAngle - file/dir size

Quickdemo

Page 5: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

5

Appears in:American Heritage Dictionary, 3rd Ed. Houghton Mifflin, 1992

Page 6: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

6

Empirical Study

Compare Treemap (borderless) and Sunburst with users performing typical file/directory-related tasks

Evaluate task performance on both correctness and time

International Journal of Human-Computer StudiesSpecial issue on Empirical Studies of InfoVis, 53(5),Nov. 2000, pp. 663-694.

With R. Catrambone, M. Guzdial,K. McDonald

Page 7: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

7

Experiment

60 participantsParticipant only works with a small or

large hierarchy in a session

Vary order across participantsSB A, TM BTM A, SB BSB B, TM ATM B, SB A

32 on small hierarchies28 on large hierarchies

Small Hierarchy(~500 files)

Large Hierarchy(~3000 files)

A B A B

Page 8: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

8

Tasks

• Identification (naming or pointing out) of a file based on size, specifically, the largest and second largest files (Questions 1-2)• Identification of a directory based on size, specifically, the largest (Q3) • Location (pointing out) of a file, given the entire path and name (Q4-7) • Location of a file, given only the file name (Q8-9)• Identification of the deepest subdirectory (Q10)• Identification of a directory containing files of a particular type (Q11) • Identification of a file based on type and size, specifically, the largest file of a particular type (Q12)• Comparison of two files by size (Q13)• Location of two duplicated directory structures (Q14)• Comparison of two directories by size (Q15)• Comparison of two directories by number of files contained (Q16)

Representative of things people really do

Page 9: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

9

Results (1)

Small hierarchy

A BPhase 1 TM 9.88 (3.23) TM 11.50 (2.14) SB 12.88 (1.96) SB 10.38 (1.69)

Phase 2 TM 12.25 (1.75) TM 10.75 (2.77) SB 12.63 (2.00) SB 11.50 (2.00)

Correct task completions (max=16)

Page 10: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

10

Results (2)

Large hierarchy

A BPhase 1 TM 8.71 (1.60) TM 8.29 (2.14) SB 11.43 (1.27) SB 11.14 (2.67)

Phase 2 TM 11.57 (1.27) TM 10.86 (1.57) SB 11.00 (2.16) SB 11.00 (2.00)

Correct task completions (max=16)

Page 11: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

11

Results (3)

Ordering/learning effect for Treemap on large hierarchies

Performance trends favored Sunburst, but not clear-cut (completions, time)

Subjective preference:SB (51), TM (9), unsure (1)

Page 12: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

12

Observations

SB appeared to convey structure better

Participants felt TM conveyed size better, but not bore out by data

Strategies mattered

Page 13: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

13

SunBurst Negative

In large hierarchies, files at the periphery are usually tiny and very difficult to distinguish

examples

Page 14: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

14

Treemap Too

examples

Page 15: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

15

Fix: Objectives

Make small slices bigger

Maintain full circular space-filling idea

Allow detailed examination of small files within context of entire hierarchy

Don’t alter ratios of sizes

Avoid use of multiple windows or lots of scrollbars

Provide an aesthetically pleasing interface in which it is easy to track changes in focus

Page 16: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

16

3 Solutions

Three visualization+navigation techniques developed to help remedy the shortcoming Angular detail Detail outside Detail inside

Proceedings of Information Visualization 2000Oct. 2000, pp. 57-65.

With E.Zhang

Page 17: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

17

Design 1 - Angular Detail

• Most “natural”• Least space-efficient• Most configurable by user

Page 18: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

18

Design 2 - Detail Outside

• Exhibits non-distorted miniature of overview• Somewhat visually disconcerting• Focus is quite enlarged (large circumference and 360°)• Relatively space efficient

Page 19: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

19

Design 3 - Detail Inside

• Perhaps least intuitive and most distorting• Items in overview are more distinct (larger circumference)• Interior 360° for focus is often sufficient

Page 20: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

20

Video

4 minutes

Page 21: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

21

Key Components

Two ways to increase area for focus region: larger sweep angle and longer circumference

Smooth transitions between overview and focus allow viewer to track changes

Always display overviewAllow focus selections from anywhere:

normal display, focus or overview regions

Page 22: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

22

Implementation

Utilizes fundamental animation update routine Example: Detail Outside (called 3 times)

Shrink global viewFocus region grows outFocus regions wraps around global view

Smooth interpolation between start-end position and angle

Page 23: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

23

Speed Considerations

Don’t draw small slicesCache small and large images of entire

hierarchy, reload rather than drawDuring animation transitions, only draw

the 100 largest slices (don’t use thresh-holding)->Consistent speed as hierarchy grows (really dependent on processor & graphics)

Page 24: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

24

Treemap Application

Can similar techniques be applied to treemaps to support focusing on small items? Expand area given to focus item Smoothly animate focus change Challenge: Where to position focus and

context?

Page 25: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

25

Idea 1: Superimpose

+: Simple- : Obscures areas

Page 26: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

26

Idea 2: Grow/Shrink

+: Shows all context- : Empty areas

Page 27: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

27

Idea 3: Shrink overview

+: Shows all context- : Wasteful of space

Page 28: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

28

Thoughts?

Other possibilities?Which do you like best?Somebody build one!

Page 29: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

29

Summary

Treemaps can be a very helpful information visualization

Techniques for examining small regions in more detail can make them even more powerful

Page 30: Musings on Treemaps John Stasko Information Interfaces Research Group College of Computing / GVU Center Georgia Institute of Technology HCIL, Univ. of

30

For More Information...

[email protected]/gvu/ii