Section checkpoint

Core Models

9 questions drawn from the 3 chapters in this section. If most land on the first try, you are ready for the next section. If not, the explanation after each answer says which idea to revisit.

Checkpoint

9 questions · first-try answers earn the most XP

1. Why can a single extreme outlier distort a fitted linear regression line noticeably?

2. Price falls with distance from the city centre in a curve rather than a straight line. Which model family suits this better?

3. Why start with linear regression even when a fancier model might score higher?

4. Why is logistic regression classified as a classification model despite its name?

5. Why does the chapter insist the decision threshold is a business decision?

6. Why prefer F1 over accuracy when classes are imbalanced?

7. Why do decision trees often beat a linear model on tabular business data without any scaling?

8. A stakeholder asks which feature matters most. What does a tree model's feature importance actually tell you?

9. Your dataset has 500 rows and 60 features, and a single tree performs worse than a forest. Why does the forest help so much here?

What this checkpoint covers

Next section: Evaluate & Ship