Given this training data: Point A (2, 3): Class = Red Point B (5, 4): Class = Blue Point C (3, 2): Class = Red Point D (6, 5): Class = Blue A new point is at (4, 3). Using K=3, which class would KNN predict? Calculate the Euclidean distances to explain your answer.

SMOTE creates ______ samples of the minority class.

In linear SVM, the coefficients are accessed via model.______[0]

What does SMOTE do?

What is the formula for a 95% confidence interval from CV?

What is the difference between Gini impurity and entropy?

When might Random Forest be preferred over XGBoost?

What does bagging primarily reduce?

What is the learning_rate parameter in gradient boosting?

You deployed a fraud detection model 6 months ago. Initially it had 95% precision, but now it has dropped to 75%. Explain two possible types of drift that could cause this degradation. What steps would you take to diagnose and address the issue?