OgbonLab

Chapter 10 Quiz: Feature Engineering

Chapter 9: Engineering Features from Geoscience Data

Learning objectives

Test your understanding of encoding, scaling, feature selection, and pipeline construction

This quiz covers both the lecture material and lab exercises from Chapter 10.

Key Concepts Review

Encoding: One-hot for nominal features; label encoding for ordinal or tree models.
Scaling: Min-max normalizes to [0,1]; standardization gives mean=0, std=1.
Feature Selection: Filter (correlation), wrapper (RFE), embedded (Lasso).
Pipeline: Ensures proper data flow and prevents leakage in cross-validation.
Interactions: $x_1 \times x_2$ captures combined effects.

References

Géron, A. (2022). Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow (3rd ed.), ch. 2. O’Reilly.
Murphy, K.P. (2022). Probabilistic Machine Learning: An Introduction, ch. 4. MIT Press.
Karpatne, A., Atluri, G., Faghmous, J.H., et al. (2017). Theory-guided data science. IEEE Trans. Knowl. Data Eng. 29(10), 2318–2331.