3. Machine Learning

Feature Engineering — Quiz

Test your understanding of feature engineering with 5 practice questions.

Practice Questions

Question 1

Which of the following feature engineering techniques is most appropriate for handling a numerical feature with a highly skewed distribution, such as 'income' or 'population', to make it more symmetrical and suitable for linear models?

Question 2

When dealing with a dataset containing a large number of categorical features, each with many unique values (high cardinality), which encoding technique is most likely to lead to the 'curse of dimensionality' and increased training time?

Question 3

Consider a scenario where you are building a predictive model for house prices. You have a feature 'Number of Rooms' and another feature 'Area of House' (in square feet). To capture the interaction between these two features, which of the following new features would be most informative?

Question 4

You are working with a dataset where a numerical feature 'Transaction Amount' has a few extreme outliers. Which of the following imputation methods for missing values would be least affected by these outliers?

Question 5

Which of the following statements accurately describes the primary difference between feature selection and feature extraction?