The key purpose of splitting the dataset into training and test sets is A) To speed up the training process 8) To reduce the amount of labelled data needed for evaluating classifier accuracy C) To reduce the number of features we need to consider as input to the learning algorithm D) To estimate how well the learned model will generalize to new/unseen data 3- k-NN algorithm can be used for A) Regression B) Classification C) Both A and B D) None of the above