25.2 Project phases

The main project phases are:

After data gathering iteration is trump

Figure from (Kuhn and Johnson 2018) (Image Credit: Owlsmcgee [Public domain] )

EDA => exploratory data analysis
source http://www.feat.engineering/intro-intro.html#the-model-versus-the-modeling-process]

  • Exploratory data analysis

    • Find correlations or mutial depence
  • Quantiative analysis
    • Check distribution
      • Long tail => log of variable
  • Feature engineering 23
    • Create and select meaningful features
  • Model fit
    • Selecting a few suited models
  • Model tuning
    • Vary model hyperpparameters

25.2.1 Feature engineering

Variables that go into model are called:

  • PredictorsSmiley face

  • Features
  • Independent variables

Quantity being modeled called:

  • PredictionSmiley face

  • Outcome
  • Response
  • Dependent variable

From input to output

\[outcome = f(features) = f(X_1, X_2, \dots, Xp) = f(X)\]

\[\hat{Y} = \hat{f}(X)\]

References

———. 2018. “Feature Engineering and Selection: A Practical Approach for Predictive Models.” http://www.feat.engineering/index.html.

  1. Good source for feature engineering: http://www.feat.engineering/index.html↩︎