(a) Give two examples of practical problems of supervised machine learning[4 marks]and identify samples and labels for them. (b) What is meant by a classification problem in machine learning? When is a classification problem called binary? When is it called multi-class? [4 marks] (c) What is meant by a regression problem in machine learning? (d) What is meant by a feature in machine learning? What is the differencebetween discrete and continuous features?[4 marks] (e) Compare and contrast batch and online learning protocols in machine learning.[6 marks] (f) Consider the following regression problem. The training set is:

The test set consists of two samples, (0, 1,0) and (0,0,0). i. Calculate the predicted labels for the test set using the K Nearest Neighbours algorithm with Euclidean distance for K = 2.[6 marks] ii. Now you are told that the true labels of the test samples (0,1,0) and(0,0,0) are 1 and 0, respectively. Calculate the test TSS, test RSS, andtest R² for your predictions. What does the value of test R² tell you aboutthe quality of the predictions?[7 marks] iii. Which method would you use to compute the test R² in scikit-learn?[1 mark]