Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
RubyConf_2017_-_Kelsey_Pedersen.pdf
Search
Kelsey Pedersen
November 17, 2017
0
29
RubyConf_2017_-_Kelsey_Pedersen.pdf
Kelsey Pedersen
November 17, 2017
Tweet
Share
More Decks by Kelsey Pedersen
See All by Kelsey Pedersen
Defying the Odds: How to Apply (and Get Selected!) to Speak at Conferences
kelseypedersen
0
39
Augmenting Human Decision Making with Data Science - PyCon
kelseypedersen
0
110
Augmenting Human Decision Making with Data Science - RubyxElixir Conf 2018
kelseypedersen
1
93
RubyConf_2017_-_Kelsey_Pedersen.pdf
kelseypedersen
0
28
Augmenting Human Decision Making with Data Science
kelseypedersen
1
89
Featured
See All Featured
Designing the Hi-DPI Web
ddemaree
280
34k
The Cost Of JavaScript in 2023
addyosmani
45
6.8k
The Cult of Friendly URLs
andyhume
78
6k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
25
1.8k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
26
1.4k
Visualization
eitanlees
145
15k
Raft: Consensus for Rubyists
vanstee
136
6.6k
Done Done
chrislema
181
16k
GraphQLの誤解/rethinking-graphql
sonatard
67
10k
Bootstrapping a Software Product
garrettdimon
PRO
305
110k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
44
6.8k
Six Lessons from altMBA
skipperchong
27
3.5k
Transcript
HUMAN DECISION MAKING Kelsey Pedersen Software Engineer, Stitch Fix with
DATA SCIENCE AUGMENTING @kelsey_pedersen
@kelsey_pedersen KELSEY PEDERSEN
I’m a software engineer at Stitch Fix - a personalized
styling service
None
None
None
None
None
None
stylists data scientists
None
None
How do humans make decisions? What are the limitations of
human decision making? How can we help users make decisions with data science?
How do humans make decisions? Human Decisions Limits of Decisions
Stylists Decisions Humans & Data Science The Future
System 1 System 2
System 1 impressions & feelings
None
None
None
None
95% of human decisions are made instantaneously
System 2 analytical & effortful
1737 x 1990 Answer: 3,456,630
None
None
System 1 gut feelings System 2 computations
LEFT BRAIN RIGHT BRAIN
What are the limits in human decision making? Human Decisions
Limits of Decisions Stylists Decisions Humans & Data Science The Future
None
Human decisions are unpredictable
Inconsistent Judgments
Inconsistent Judgments
Inconsistent Judgments
Human decisions are driven by your own past experiences
Limited Information
Limited Information
Limited Information
Human decisions are driven by your own views and preferences
Biased Decisions
Biased Decisions
Biased Decisions
RISKY & UNPREDICTABLE
What are examples of Stitch Fix stylists making decisions? Human
Decisions Limits of Decisions Stylists Decisions Humans & Data Science The Future
Inconsistent Judgments by Stylists
Inconsistent Judgments by Stylists
Limited Information by Stylists
Limited Information by Stylists
Biased Decisions by Stylists
None
In what ways can data science augment human decisions? Human
Decisions Limits of Decisions Stylists Decisions Humans & Data Science The Future
None
Data Science Noun: Use of mathematics or statistics to answer
a business question
GUIDE DECISIONS WITH COMPUTATIONS TRAIN DECISIONS WITH FEEDBACK
DATA STYLING STYLING STYLING STYLING STYLING STYLING STYLING STYLING DATA
DATA DATA DATA DATA
None
Deciding on one item
None
72% 60% 48% 44%
None
None
None
Deciding on all 5 items
72% 60% 48% 44% 76%
72% 60% 48% 44% 24%
None
Client Feedback on all 5 items
None
None
None
Client Feedback overtime for one Stylist
None
None
None
All Feedback for all Stylists
None
None
Can humans augment algorithms? Human Decisions Limits of Decisions Stylists
Decisions Humans & Data Science The Future
None
None
None
None
None
72% 60% 48% 10%
72% 60% 48% 44% 24%
None
None
stylists data scientists
What is the future of data science? Human Decisions Limits
of Decisions Stylists Decisions Humans & Data Science The Future
None
LEFT BRAIN RIGHT BRAIN VS.
System 1 gut feelings System 2 computations &
None
None
Data Science Humans System 1 System 2 Data Science Humans
System 3
System 3 predictive & intuitive
stylists data scientists
TRAIN Performance Metrics
None
System 1 gut feelings System 2 computations System 3 feedback
What is the business impact of system 3?
lower labor cost fewer mistakes from humans greater client satisfaction
increased keep rate
In conclusion…
Conclusion
None
None
Conclusion
humans data science
@kelsey_pedersen KELSEY PEDERSEN