← 03 Data Prep & Feature Engineering
Area 03 test · Core ML

Data Prep & Feature Engineering

20 questions · 70% to pass · no time limit. You see the right answers and why after you submit.

  1. 1What should be your first step with a new dataset?
  2. 2Which column is target leakage for a churn model?
  3. 3Why drop customer_id as a feature?
  4. 4A right-skewed numeric column has missing values. Best simple fill?
  5. 5When is a missing indicator most useful?
  6. 6Q1 = 20, Q3 = 40. The IQR upper fence is…
  7. 7A customer spends 100× the average and the value is real. Reasonable treatment?
  8. 8A 'region' column with north/south/east/west for logistic regression should be…
  9. 9Ordinal encoding is appropriate for…
  10. 10What does OneHotEncoder(handle_unknown="ignore") do with a category never seen in training?
  11. 11Which model does NOT need feature scaling?
  12. 12After StandardScaler on training data, each column has…
  13. 13Why might min-max scaling be a poor choice with outliers?
  14. 14What is wrong with fitting a scaler on all rows and then splitting?
  15. 15The correct order is…
  16. 16Churn is 30% of rows. Why stratify the split?
  17. 17You predict next month's sales from two years of daily data. How should you split?
  18. 18Why put preprocessing inside a scikit-learn Pipeline for cross-validation?
  19. 19Several rows belong to the same customer. Which splitter avoids that customer appearing in both train and test?
  20. 20In the lab, removing refund_issued drops the test AUC from about 0.94 to about 0.77. What does that mean?
0 of 20 answered