Replace cluster means with representative observations

Specify this option to create a column in the output data set containing one observation from each cluster. The observation selected is the one with the smallest distance to the cluster mean. This option guarantees that original variables are still used for prediction.

Note: This option applies only to the non-interactive clustering methods.

To Specify This Option:

Select either Automated K Means or Automated Radius K Means as the Clustering Method to enable the Replace Cluster Means with Representative Observations check box.

Click within the check box.