Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Augmenting Human Decision Making with Data Scie...

Sponsored · SiteGround - Reliable hosting with speed, security, and support you can count on. →
Avatar for Kelsey Pedersen Kelsey Pedersen
May 07, 2018
Technology
100
1

Augmenting Human Decision Making with Data Science - RubyxElixir Conf 2018

Avatar for Kelsey Pedersen

Kelsey Pedersen

May 07, 2018

More Decks by Kelsey Pedersen

See All by Kelsey Pedersen
Defying the Odds: How to Apply (and Get Selected!) to Speak at Conferences
Avatar for Kelsey Pedersen kelseypedersen
0
51
Augmenting Human Decision Making with Data Science - PyCon
Avatar for Kelsey Pedersen kelseypedersen
0
110
RubyConf_2017_-_Kelsey_Pedersen.pdf
Avatar for Kelsey Pedersen kelseypedersen
0
36
RubyConf_2017_-_Kelsey_Pedersen.pdf
Avatar for Kelsey Pedersen kelseypedersen
0
39
Augmenting Human Decision Making with Data Science
Avatar for Kelsey Pedersen kelseypedersen
1
100

Other Decks in Technology

See All in Technology
地元にいないローカルオーガナイザーの立ち回り
Avatar for uvb_76 uvb_76
1
400
Terraformモジュールは、なぜ「魔境」化するのか
Avatar for hayama hayama17
1
130
Agentic AI時代における メルカリのAIガバナンスとガードレール実装
Avatar for ichihara naoichihara
17
17k
大学生が本気でDatabricksを活用してDiscordサークルをデータ駆動させてみた
Avatar for Kazuki Date phantomjuju
1
310
自称宇宙最速で不合格となったAIP-C01にリベンジを果たすべくAIで問題集アプリを作ってみた。
Avatar for Yuuki Yamashita yama3133
0
250
Databricks 月刊サービスアップデート 2026年05月号
Avatar for ちょし tyosi1212
0
130
ルールやカスタム機能、どう使う？理想の出力を引き出すために今知りたいIBM Bob 5つの機能
Avatar for mu (Mizuho Uehara) muehara
0
160
Claude Codeですべての日常業務を爆速化しよう！
Avatar for みのるん minorun365
PRO
17
16k
大規模災害時でも高い信頼性を維持するアプリケーション基盤の実現/nikkei-tech-talk46
Avatar for 日本経済新聞社 エンジニア採用事務局 nikkei_engineer_recruiting
0
130
開発を止めない CI/CD ~CI Visibilityによる継続的最適化~
Avatar for Ryosuke Tsuruda pensuke628
0
220
Platform engineering for developers, architects & the rest of us (AI agents)
Avatar for danielbryantuk danielbryantuk
0
160
oracle-to-databricks-migration-with-llm-and-dbt
Avatar for case-k-git casek
1
390

Featured

See All Featured
Sharpening the Axe: The Primacy of Toolmaking
Avatar for Bryan Cantrill bcantrill
46
2.8k
The #1 spot is gone: here's how to win anyway
Avatar for Tamara Novitovic tamaranovitovic
2
1.1k
Optimising Largest Contentful Paint
Avatar for Harry Roberts csswizardry
37
3.7k
CSS Pre-Processors: Stylus, Less & Sass
Avatar for Bermon Painter bermonpainter
360
30k
Performance Is Good for Brains [We Love Speed 2024]
Avatar for Tammy Everts tammyeverts
12
1.7k
Discover your Explorer Soul
Avatar for Emna emna__ayadi
2
1.1k
Embracing the Ebb and Flow
Avatar for Simon Collison colly
88
5.1k
エンジニアに許された特別な時間の終わり
Avatar for watany watany
107
240k
Marketing Yourself as an Engineer | Alaka | Gurzu
Avatar for Gurzu gurzu
0
210
ラッコキーワード サービス紹介資料
Avatar for ラッコ株式会社 rakko
1
3.5M
The Invisible Side of Design
Avatar for Vitaly Friedman smashingmag
302
52k
Highjacked: Video Game Concept Design
Avatar for Ryan Kendrick rkendrick25
PRO
1
370

Transcript

  1. HUMAN DECISION MAKING Kelsey Pedersen Software Engineer, Stitch Fix with

    DATA SCIENCE AUGMENTING @kelsey_pedersen
  2. None
  3. None
  4. None
  5. None
  6. None
  7. None
  8. None

  9. @kelsey_pedersen

  10. None
  11. None

  12. The future of software development is the integration of algorithms

  13. None

  14. How do humans make decisions? What are the limitations of

    human decision making? How can we help users make decisions with data science?

  15. How do humans make decisions? Human Decisions Limits of Decisions

    Stylists Decisions Humans & Data Science The Future

  16. System 1 System 2 FAST SLOW

  17. System 1 impressions & feelings

  18. None
  19. None
  20. None
  21. None

  22. 95% of human decisions are made instantaneously

  23. System 2 analytical & effortful

  24. 1737 x 1990 Answer: 3,456,630

  25. None

  26. System 1 gut feelings System 2 computations

  27. What are the limits in human decision making? Human Decisions

    Limits of Decisions Stylists Decisions Humans & Data Science The Future
  28. None

  29. Human decisions are unpredictable

  30. None
  31. None

  32. Human decisions are driven by your own past experiences

  33. None

  34. Human decisions are driven by your own views and preferences

  35. None
  36. None

  37. What is an example of balancing humans and data science?

    Human Decisions Limits of Decisions Stylists Decisions Humans & Data Science The Future
  38. None

  39. Inconsistent Judgments by Stylists

  40. Information Overload by Stylists

  41. Biased Decisions by Stylists

  42. None

  43. In what ways can data science augment human decisions? Human

    Decisions Limits of Decisions Stylists Decisions Humans & Data Science The Future

  44. GUIDE DECISIONS WITH COMPUTATIONS TRAIN DECISIONS WITH FEEDBACK

  45. None

  46. Deciding on one item

  47. None

  48. 72% 60% 48% 44%

  49. None
  50. None
  51. None

  52. Deciding on all 5 items

  53. 72% 60% 48% 44% 56%

  54. 50% 26% 18% 14% 27%

  55. None
  56. None

  57. Client Feedback on all 5 items

  58. None
  59. None
  60. None

  61. Client Feedback overtime for one Stylist

  62. None
  63. None
  64. None

  65. All Feedback for all Stylists

  66. None
  67. None

  68. Can humans augment algorithms? Human Decisions Limits of Decisions Stylists

    Decisions Humans & Data Science The Future
  69. None
  70. None
  71. None
  72. None
  73. None

  74. 72% 60% 48% 10%

  75. 72% 60% 48% 44% 24%

  76. None
  77. None

  78. stylists data scientists

  79. What is the future of humans and data science? Human

    Decisions Limits of Decisions Stylists Decisions Humans & Data Science The Future
  80. None

  81. System 1 gut feelings System 2 computations &

  82. None
  83. None

  84. Data Science Humans System 1 System 2 Data Science Humans

    System 3

  85. System 3 predictive & intuitive

  86. TRAIN Performance Metrics

  87. System 1 gut feelings System 2 computations System 3 feedback

  88. What is the business impact of system 3?

  89. lower labor cost fewer mistakes from humans greater client satisfaction

    increased keep rate

  90. In conclusion…

  91. Conclusion

  92. None
  93. None

  94. Conclusion

  95. humans data science

  96. @kelsey_pedersen KELSEY PEDERSEN