Question 7
UnclassifiedPolynomial features are used to:
Correct answer: B
Explanation
Polynomial features transform inputs by adding "interaction and power terms," which lets a linear model represent curved or more complex patterns. This enables linear models to capture non-linear relationships while still using a linear fitting method on the expanded feature set.
Why each option is right or wrong
A. Reduce model complexity by removing variables and simplifying the feature space.
B. Enable linear models to capture non-linear relationships by adding interaction and power terms.
Polynomial feature expansion creates additional terms such as x^2, x^3, and cross-products like x1x2, so a model that is linear in its coefficients can fit curved decision boundaries or response surfaces. In practice, the original linear estimator is then trained on this expanded design matrix, allowing it to represent non-linear structure without changing the underlying fitting algorithm.
C. Improve training speed by decreasing the number of computations required.
D. Eliminate multicollinearity by making all predictors independent of one another.