Tools

To study and implement the K-Means Clustering algorithm for similarity-based grouping

  1. Click on the CREATE tab.
  2. Enter the values of X and Y in the attributes section and click on the ADD button to add values to the table.
  3. Repeat steps 2 until there are 6 data-points in the table.
  4. Click on the NEXT button to specify the number of clusters (K) you want to form, select the value of clusters, click on the SUBMIT button, and then again click on the NEXT button.
  5. Now, initialize centroids by selecting randomly K centroids and clicking on the SUBMIT button. Afterward, click on the NEXT button.
  6. Find the euclidean distance between a centroid and a data point by clicking on the buttons below the table.
  7. Now, determine the cluster assignment for the data point. To proceed, click on the highlighted cell (K) in the table and enter the corresponding cluster number. Hint: Provide the position of the minimum value (e.g., 1.)
  8. Repeat steps 7 & 8 for all the data points in the table.
  9. Click on the CENTROIDS button to compute the new centroid of each cluster.
  10. Repeat steps 6 to 9 until no data points change clusters.
  11. If the old cluster matches the new cluster, click on the NEXT button to proceed further.
  12. Finally, click on the PLOT button to display the graph.