Question
Can someone please help me with this problem using exel? I'm really struggling Directions: Please feel free to collaborate on the conceptualization of this exercise. Nonetheless, the implementation and solution of this exercise must be done individually for full credit The K-Means Algorithm (Lloyd, 1982) is a clustering technique that helps us understand unsupervised clustering processes The algorithm partitions the data into disjoint clusters (or classes). The algorithm is as follows: . Choose a value for K number of clusters. We arbitrarily choose K-2 for this exercise 2Choose K instances idata points) within the dataset at random. These are the initial cluster centers. We chose instance 1(1.0,1.5) as the first cluster center Ch, and instance 21.0.45)as the second cluster centerG Use simple Euclidean distance: Distance (A-B)-VX-XP+(Y-Yo to assign nstances to their clusters. Choose the minimum distance to make the assignment Calculate the sum of the square of these minimum distances (SSE: sum of squared eror) as an indicator of quality 4. Use the instances within each cluster to compute the new center for each cluster using a simple average (means) of the x coordinates and y coordinates for the member instances lfthe new mean values are identical to the mean values ofthe previous iteration clusters. the process terminates. That is, the algorithm terminates after several iterations if for each cluster all instances found to belong in that cluster were also the same ones from the previous iteration. Otherwise,use the new means as cluster centers and repeat stops3 through 5. 5 This is data from -An Example Using K-Means"found on page 85 of your textbook Table 3.6 . K-Means Input Values Instance 1.5 4.5 1.5 3.5 25 60 20 20 3.0 Ihave started an Excel spreadshet to be used as a template to help you bring this exercise to ompletion. Your final answer should be C(1.8,2.7) and C:(50,60) with instances 1, 2,3,4 and 5 in cluster 1 and instance 6 alone in cluster 2 with a SSE-9.60
Explanation / Answer
Cluster and there position K 2