Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Visualization for Discovery

682 views

Published on

Presenter: Prof. Jeffrey Heer, University of Washington & Trifacta Inc.

Published in: Technology
  • Be the first to comment

Visualization for Discovery

  1. 1. Jeffrey Heer @jeffrey_heer Univ. of Washington + Trifacta Visualization for DISCOVERY
  2. 2. Effectiveness of Penicillin, Neomycin & Streptomycin vs. Bacteria Species
  3. 3. Which antibiotic should one use?
  4. 4. Show data variation, not design variation. - Edward Tufte
  5. 5. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other Wainer & Lysen Am. Sci. 2009
  6. 6. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other Wainer & Lysen Am. Sci. 2009 What does antibiotic response reveal about the biology of bacteria?
  7. 7. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other Wainer & Lysen Am. Sci. 2009
  8. 8. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other Wainer & Lysen Am. Sci. 2009 ? ?
  9. 9. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other Wainer & Lysen Am. Sci. 2009 Not a streptococcus! Actually a streptococcus!
  10. 10. How might our tools spur new questions and prompt skepticism?
  11. 11. 0 10 20 30 40 Neomycin 0 200 400 600 800 Penicillin
  12. 12. 0.001 0.01 0.1 1 10 100 Log10(Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(Penicillin)
  13. 13. 0.0010.010.1110100 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin)
  14. 14. 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin)
  15. 15. 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Aerobacter Brucella Bacillus Diplococcus Escherichia Klebsiella Mycobacterium Proteus Pseudomonas Salmonella Staphylococcus Streptococcus
  16. 16. 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other
  17. 17. Aerobacter aerogenes Brucella abortus Bacillus anthracis Diplococcus pneumoniae Escherichia coli Klebsiella pneumoniae Mycobacterium tuberculosis Proteus vulgaris Pseudomonas aeruginosa Salmonella typhosa Salmonella schottmuelleri Staphylococcus albus Staphylococcus aureus Streptococcus fecalis Streptococcus hemolyticus Streptococcus viridans 0.0010.010.11101001,000 Log10(1 / Neomycin) 0.001 0.01 0.1 1 10 100 1,000 Log10(1/Penicillin) Genus Streptococcus Staphylococcus Salmonella Other
  18. 18. A Combinatorial Design Space
  19. 19. 1. Variable Selection A Combinatorial Design Space
  20. 20. 1. Variable Selection 2. Data Transformation A Combinatorial Design Space
  21. 21. 1. Variable Selection 2. Data Transformation 3. Visual Encoding Design A Combinatorial Design Space
  22. 22. 1. Variable Selection 2. Data Transformation 3. Visual Encoding Design —> Thousands of possible charts! A Combinatorial Design Space
  23. 23. How might we augment manual chart construction with interactive browsing of recommended visualizations?
  24. 24. ggplot(diamonds, aes(x=price, fill=cut)) + geom_bar(position="dodge")
  25. 25. 1. Trifacta Visual Profiler 2. Data Voyager (UW + Tableau)
  26. 26. 1. Trifacta Visual Profiler 2. Data Voyager (UW + Tableau)
  27. 27. Trifacta Visual Profiler
  28. 28. OverviewOverview Trifacta Visual Profiler
  29. 29. DetailsDetails Trifacta Visual Profiler
  30. 30. Trifacta Visual Profiler
  31. 31. Trifacta Visual Profiler
  32. 32. Trifacta Visual Profiler
  33. 33. Trifacta Visual Profiler
  34. 34. Trifacta Visual Profiler
  35. 35. Trifacta Visual Profiler
  36. 36. 1. Trifacta Visual Profiler 2. Data Voyager (UW + Tableau)
  37. 37. User
  38. 38. Voyager 
 Visualization Browser User 
 Data Set
  39. 39. Voyager 
 Visualization Browser Compass Recommendation Engine Data Schema & Statistics 
 User
  40. 40. Voyager 
 Visualization Browser Compass Recommendation Engine Data Schema & Statistics 
 User 1. Select data variables 2. Apply transformations 3. Pick visual encodings
  41. 41. Voyager 
 Visualization Browser Compass Recommendation Engine Data Schema & Statistics 
 User Constrain & rank choices by data type, statistics & perceptual principles.
  42. 42. Voyager 
 Visualization Browser Data Schema & Statistics 
 Ranked and Clustered Vega-lite Specifications User Compass Recommendation Engine
  43. 43. Voyager 
 Visualization Browser Compass Recommendation Engine Vega-lite Compiler Vega-lite Specifications Ranked and Clustered Vega-lite Specifications User Data Schema & Statistics 

  44. 44. Voyager 
 Visualization Browser Compass Recommendation Engine Vega Renderer Vega-lite Compiler Vega-lite Specifications Vega
 Specifications Ranked and Clustered Vega-lite Specifications User Data Schema & Statistics 

  45. 45. Voyager 
 Visualization Browser Compass Recommendation Engine Vega Renderer Interactive Visualizations Vega-lite Specifications Vega
 Specifications Ranked and Clustered Vega-lite Specifications User Data Schema & Statistics 
 Vega-lite Compiler
  46. 46. Voyager 
 Visualization Browser Interactive Visualizations Compass Recommendation Engine Vega Renderer Interactive Visualizations Vega-lite Specifications Vega
 Specifications Ranked and Clustered Vega-lite Specifications User Data Schema & Statistics 
 Vega-lite Compiler
  47. 47. Voyager 
 Visualization Browser Interactive Visualizations User 
 Selection Compass Recommendation Engine Vega Renderer Interactive Visualizations Vega-lite Specifications Vega
 Specifications User Selection, Data Schema & Statistics 
 Ranked and Clustered Vega-lite Specifications User Vega-lite Compiler
  48. 48. Voyager 
 Visualization Browser Interactive Visualizations User 
 Selection Compass Recommendation Engine Vega Renderer Interactive Visualizations Vega-lite Specifications Vega
 Specifications User Selection, Data Schema & Statistics 
 Ranked and Clustered Vega-lite Specifications User Improves data coverage! +3x variable sets shown +1.5x more interacted with Vega-lite Compiler
  49. 49. Refining visualization recommendation What to optimize? How to evaluate? Scaling interactive visualizations Large D harder than large N… Help avoid statistical pitfalls Recognize mix effects, convey uncertainty Ongoing Challenges
  50. 50. How might our tools spur new questions and prompt skepticism?
  51. 51. vega.github.io
  52. 52. Jeffrey Heer @jeffrey_heer Univ. of Washington + Trifacta Visualization for DISCOVERY

×