Chapter 14: Conclusion
Fundamentals
general concepts
how data science fits in business
ways of thinking analytically
attract, structure, grow teams
gives competitive advantage
tactical principles
help to gather appropriate data
consider appropriate methods
data mining process
high-level data science tasks
recall business problem always
data is an asset
associate cost and benefits
constraints
applying data science to problem
generalization and overfitting
Extracting the data
identifying informative attributes
fitting a numeric function model
controlling complexity
calculate similarities b/w objects in data
Recall from book
data science strategies
structuring business problems
use expected value
decompose problems in tasks
lift
how likely a pattern is
never just say ok
data scientist ask questions
Example
Applying data science
consumers switching to mobile devices from desktops
how co's reach them on mobiles
notice availability of info on location
ask the question!
how might we use this data
Changing the way we thing
don't change the problem
happens when try to fit what the data shows
what data can't do
humans and computers are better at diff. things
data science
integrate human and computer techniques
humans
good at evaluation stage
idea to optimize objective
using both to understand wants of people
ethical issues
privacy and data issues
raises the question: what is data privacy?
Final thoughts
data scientist should be able to describe process to anyone
be wary of jargon