Download as PDF, PPTX
Uploaded bychris wiggins
data science @NYT ; inaugural Data Science Initiative Lecture
The document discusses the evolution and significance of data science, particularly as it relates to the New York Times and various historical figures like John Tukey. It highlights the impact of modern technology and methodologies on data analysis, emphasizing the changing mindsets and toolsets in the field. Additionally, it outlines essential skills, learnings, and roles that contribute to effective data science practices.
More Related Content
Similar to data science @NYT ; inaugural Data Science Initiative Lecture
data science @NYT ; inaugural Data Science Initiative Lecture
- 1. data science @The New York Times chris.wiggins@columbia.edu chris.wiggins@nytimes.com @chrishwiggins references: bit.ly/brown-refs
- 2. data science @The New York Times
- 3. data science @The New York Times
- 4.
- 5.
- 6. data science: mindset& toolset drew conway, 2010
- 7.
- 8.
- 9.
- 10.
- 11.
- 12.
- 13. B.A. & M.Sc.from Brown
- 14.
- 15. “By the endof late 1945, I was a statistician rather than a topologist”
- 16.
- 17.
- 18.
- 19. “the progenitor ofdata science.” - @mshron
- 20. “The Future ofData Analysis,” 1962 John W. Tukey
- 21.
- 22. Tukey 1965, viaJohn Chambers
- 23. TUKEY BEGAT SWHICH BEGAT R
- 24.
- 25. Tukey 1975 In 1975,while at Princeton, Tufte was asked to teach a statistics course to a group of journalists who were visiting the school to study economics. He developed a set of readings and lectures on statistical graphics, which he further developed in joint seminars he subsequently taught with renowned statistician John Tukey (a pioneer in the field of information design). These course materials became the foundation for his first book on information design, The Visual Display of Quantitative Information
- 26.
- 27.
- 28.
- 29. fast forward ->2001
- 30. “The primary agentsfor change should be university departments themselves.”
- 31. data science @The New York Timeshistories 1. slow burn @Bell: as heretical statistics (see also Breiman) 2. caught fire 2009-now: as job description historical rant: bit.ly/data-rant
- 32.
- 33. biology: 1892 vs.1995 biology changed for good.
- 34. biology: 1892 vs.1995 new toolset, new mindset
- 35. genetics: 1837 vs.2012 ML toolset; data science mindset
- 36.
- 37. genetics: 1837 vs.2012 ML toolset; data science mindset arxiv.org/abs/1105.5821 ; github.com/rajanil/mkboost
- 38. data science: mindset& toolset
- 39.
- 40.
- 41.
- 42.
- 43.
- 44.
- 45. news: 21st century churchstate engineering
- 46. 1851 1996 newspapering: 1851vs. 1996
- 47. example: millions of viewsper hour2015
- 49. "...social activities generatelarge quantities of potentially valuable data...The data were not generated for the purpose of learning; however, the potential for learning is great’’
- 50. "...social activities generatelarge quantities of potentially valuable data...The data were not generated for the purpose of learning; however, the potential for learning is great’’ - J Chambers, Bell Labs,1993
- 51.
- 52. data science: theweb is your “online presence”
- 53. data science: theweb is a microscope
- 54. data science: theweb is an experimental tool
- 55. 1851 1996 newspapering: 1851vs. 1996 vs. 2008 2008
- 56. “a startup isa temporary organization in search of a repeatable and scalable business model” —Steve Blank
- 57. every publisher isnow a startup
- 58. every publisher isnow a startup
- 60. news: 21st century churchstate engineering
- 61. news: 21st century churchstate engineering
- 62.
- 63. learnings - predictive modeling -descriptive modeling - prescriptive modeling
- 64. (actually ML, shhhh…) -(supervised learning) - (unsupervised learning) - (reinforcement learning)
- 65. learnings - predictive modeling -descriptive modeling - prescriptive modeling cf. modelingsocialdata.org
- 66. predictive modeling, e.g., cf.modelingsocialdata.org
- 67. predictive modeling, e.g., “thefunnel” cf. modelingsocialdata.org
- 68. interpretable predictive modeling supercoolstuff cf.modelingsocialdata.org
- 69. interpretable predictive modeling supercoolstuff cf.modelingsocialdata.org arxiv.org/abs/q-bio/0701021
- 70. optimization & learning,e.g., “How The New York Times Works “popular mechanics, 2015
- 71. optimization & prediction,e.g., “How The New York Times Works “popular mechanics, 2015 (some models) (somemoneys)
- 72. recommendation as predictivemodeling
- 73. recommendation as predictivemodeling bit.ly/AlexCTM
- 74. descriptive modeling, e.g, cf.daeilkim.com ; import bnpy
- 75.
- 76. modeling your audience (optimization,ultimately)
- 77. also allows insight+targetingas inference modeling your audience
- 78.
- 79.
- 80. prescriptive modeling aka “A/Btesting”; RCT cf. modelingsocialdata.org
- 81.
- 82.
- 83.
- 84.
- 85.
- 86. common requirements in datascience:
- 87. common requirements in datascience: 1. people 2. ideas 3. things cf. John Boyd, USAF
- 88.
- 89. data skills data scienceand… - data engineering - data embeds - data product - data multiliteracies cf. “data scientists at work”, ch 1
- 90. data science: ideas -new mindset > new toolset
- 91.
- 92. thanks to thedata science team!
- 93. data science @The New York Times chris.wiggins@columbia.edu chris.wiggins@nytimes.com @chrishwiggins references: bit.ly/brown-refs