Correlation Radius for Clustering

Use this feature to specify a correlation value between 0.7 and 1 to use to determine the minimum radius when Automated Radius K Means is selected as the K-Means Clustering Method.

The higher the correlation, the more clusters that are defined, and the more highly members of each cluster correlate with one another. Denoting the correlation that you specify as Rho, the radius for SAS PROC FASTCLUS is set equal to sqrt(2 (Mean_USS)(1 - rho)), where Mean_USS is the average of the columnwise uncorrected sums of squares.

Note: This option is available only when the Use K-Means clustering to reduce marker number check box is checked.

To Specify a Correlation Radius:

8

Check the Use K-Means clustering to reduce marker number check box.

8

Select Automated Radius K Means as the K-Means Clustering Method.

8

Either enter the desired value into the text field, or move the slider to the right or to the left to change the value.

Tip: To change the scale of the slider, right-click on the slider and select Rescale Slider from the pop-up menu. Change the upper and/or lower boundaries in the window that appears, to rescale the slider and click OK.