Ch 3
"data is always predictive"
data effect
technology that learns from experience (data) to predict the future behavior of individuals
predictive analytics
this was the name of the competition created by Netflix to improve their recommendations to customers
Netflix Price
the name of the AI that shares its name with a famous sidekick
Watson
use this symbol to add a comment in RStudio
#
this measures the strength of a linear relationship between two variables
correlation
this breaks down data into smaller and smaller groups
decision trees
this was the amount of teams that made up BellKor's Pragmatic Chaos
3
"What is Toronto?"
you use this function to plot data
ggplot
these ranges indicate a strong linear relationship
-1 to -0.8 and 0.8 to 1
this is the result of overtraining a model
overfitting
this was the practice Netflix used to get competitors with different perspectives and backgrounds rather than experts in predictive analytics specifically
crowdsourcing
this was the year the infamous "man vs machine" Jeopardy! game took place
2011
this operator is used to chain multiple functions together
%>% (pipe)
"there's always so much more today than yesterday"
"big data"
this is a type of model that considers multiple factors about an individual
multivariate model
this combines the strengths and weaknesses from models to create one supermodel
ensemble effect/ensemble model
using this score, Watson would either play more aggressive or more conservatively
confidence score
you use this symbol as an easier/quicker way to subset
$
emotions weren't thought to be able to be measured by predictive analytics until this...
social media
art (human creativity) drives machine learning
induction effect
this was the data Netflix gave to the competitors to use for the competition
movie ratings
the name of the project IBM's Watson was a part of
DeepQA
you use this function to force a linear relationship (linear regression)
geom_smooth(method = lm)