Encoding Techniques Quiz

5 questions Pass: 70% +25 pts (50 with Pro)

Quiz covering Feature Engineering

What you'll be asked

  1. Which encoding technique is most appropriate for a categorical variable with no inherent order, such as 'Color' (Red, Blue, Green)?
  2. When using Label Encoding on an ordinal feature like 'Education Level' (High School < Bachelors < Masters), why is it preferred over One-Hot Encoding?
  3. You are dealing with a categorical feature containing over 1,000 unique values (high cardinality). Which technique should you avoid to prevent the 'curse of dimensionality'?
  4. What is the primary risk of using Target Encoding (or Mean Encoding) on a small dataset?
  5. You have a categorical feature with high cardinality and you need a memory-efficient method that doesn't store a mapping dictionary. Which technique is best suited?

Answers and explanations come after you submit.