Module 3 Assessment: Distance-Based Methods and Kernel Intuition Check

Estimated active time: 30-45 minutes

Question 1

How does the score change when features are left unscaled, and why?

Pass answer: gives both figures and identifies the column whose range dominates the distance.

Question 2

What does k trade off in your results?

Pass answer: cites specific scores at different k and describes the bias-variance movement they show.

Question 3

Why is k-nearest neighbours expensive at prediction time rather than training time?

Pass answer: recognises that there is no fitted model — every query searches the training set.

Question 4

What simpler baseline must this beat, and does it?

Pass answer: compares against both Module 1 baselines with figures.