Mahalanobis measure is one of many distance measure used for k-means. Given the below definition, describe the shape of the covariance matrix correspond- ing to o, and explain why Mahalanobis measure is also called scaled euclidean measure in this case. d(x, c): = (x - C) $ (b) [8 pts] Given a dataset x = {0, 2, 4, 6, 18, 20}, initialize the k-means clustering algorithm with 2 cluster centers c = 3 and c 4. What are the values of c and c after the first iteration of k-means? Also report the values after the second iteration. (c) [8 pts] Given the same dataset as in (b), perform greedy initialization to get the initial k = 3 centers. Start with c 4. Below is the greedy initialization process. = a) Choose c b) Choose the next center c; to be argmaxex {D(x)} D(x) = min ||xc|| CEC c) Repeat step b) until k centers are chosen where at any given time, with the current set of cluster centers C,