IBM Data WH K-Means Nugget - Model Tab

The Model tab contains various graphic views that show summary statistics and distributions for fields of clusters. You can export the data from the model, or you can export the view as a graphic.

When you are working with IBM® Netezza® Analytics Version 2.x or previous, or when you build the model with Mahalanobis as the distance measure, the content of the K-Means models is shown in textual format only.

For these versions, the following information is shown:

• Summary Statistics. For both the smallest and the largest cluster, summary statistics shows the number of records. Summary statistics also shows the percentage of the data set that is taken up by these clusters. The list also shows the size ratio of the largest cluster to the smallest.
• Clustering Summary. The clustering summary lists the clusters that are created by the algorithm. For each cluster, the table shows the number of records in that cluster, together with the mean distance from the cluster center for those records.