Feature Engineering and Selection Flashcards
7 cards from real Data Science practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Feature Engineering and Selection flashcards as text
Which feature selection method evaluates subsets of features using the actual model's performance?
Answer: Wrapper methods
Wrapper methods (e.g., recursive feature elimination) train the model to score feature subsets.
Filter methods select features based on:
Answer: Statistical measures independent of any model
Filter methods rank features by statistics like correlation or chi-square, independent of a model.
L1 (Lasso) regularization performs feature selection by:
Answer: Shrinking some coefficients exactly to zero
L1 penalty can drive coefficients to exactly zero, effectively removing those features.
Two features have a correlation of 0.97. What is a common action in feature selection?
Answer: Drop one of the redundant features
Highly correlated features are redundant, so dropping one reduces multicollinearity.
Which metric measures multicollinearity among features?
Answer: Variance Inflation Factor (VIF)
VIF quantifies how much a feature is linearly explained by the others.
Recursive Feature Elimination (RFE) works by:
Answer: Iteratively removing the least important features and refitting
RFE repeatedly trains the model and prunes the weakest features until the target count remains.
A filter method using mutual information captures:
Answer: Both linear and non-linear dependency with the target
Mutual information measures any statistical dependency, including non-linear relationships.