TOPIC: Eucilidean distance

Question No. 3: [08 Marks] Suppose that the data mining task is to cluster points (with (x, y) representing location) into four clusters, where the points are: A1(3,6). A2(2,5). A3(3,9), B1(4,8), B2(1,4), B3(5.6), C1(7,3), C2(9.4), C3(8,2), D1(3,7). D2(1,9), D3(8,1). The distance function is Euclidean distance. Suppose initially we assign Al, B2, C3, and D2 as the center of each cluster, respectively. Use the k-means algorithm to show: a) The four cluster centers after the first round of execution. b) The final four clusters. c) Visually represent the clusters of both parts (a) and (b) d) Normalize the values of the input variables to adjust for the different magnitudes of the variables. How many clusters do you recommend? Why? Rubrics: Understanding & Formulate = 2 Marks Steps = 3 Marks Problem Solving = 2 Marks Correct Output = 1 Mark