Phase 2: Machine Learning Fundamentals (Questions 11-18)

Question 11: You need to predict the probability of diabetes based on age and body fat percentage. Which ML model should you use?

Rationale: TWO features (age, body fat) predicting a numeric value (probability) = Multiple linear regression.

Question 12: Which type of machine learning algorithm predicts a numeric label based on item features?

Rationale: Predicting numeric values = Regression.

Question 13: A retailer wants to group customers with similar attributes for targeted marketing. Which ML type is this?

Rationale: Grouping similar items without predefined labels = Clustering (unsupervised).

Question 14: Which ML algorithm finds optimal groupings WITHOUT relying on labeled training data?

Rationale: No labels required = Unsupervised learning = Clustering.

Question 15: For an e-scooter rental prediction model, which attributes are FEATURES?

Weather includes: temperature, weekday/weekend Predictions: battery levels, distance traveled, number of hires

Rationale: Features = inputs (temperature, day type). Labels = predictions (hires, battery, distance).

Question 16: Which Azure Machine Learning Designer module is used to TRAIN a model?

Rationale: Algorithm modules (Linear Regression) train models. Clean/Select are data prep. Evaluate measures performance.

Question 17: Which assumption must multiple linear regression satisfy to avoid misleading predictions?

Rationale: Features must be independent (no multicollinearity).

Question 18: What should a model include to predict diabetes probability from age and body fat percentage?

Rationale: Age and body fat = 2 features. Diabetes probability = 1 label.