Hi Doug Thank You for the detailed explanation of the problem.   "Identify meaningful subsets of variables to create clusters.  You can then use the 'cluster' built on each subset of variables to create a profile which uses many dimensions -- one for each subset.  This is much easier to interpret than an overall cluster solution.     * Identify subsets of observations which might not naturally group neatly (e.g. high-value customers vs. low-value customers) and then cluster the larger groups separately to allow for different variables to be used in profiling."   I really liked these points of yours... and that's what I'm doing now.     Cheers Roby ... View more

Hi Terry,   Greatly appreciate your explanation and the solution template. Your solution definitely saves time in modifying the labels of the profile plots.Thank You.   Sorry if I was not clear enough in the initial query, rather than replacing or modifying the labels of the training data what I was looking for was the ability to globally visualise the labels within the Segment Size pie chart on the Result pane of the Segment Profile node, without me needing to double click on each input profile for every segment and then right click to display the legend.   For the segment profile (attached) I double-clicked on over 60 input profiles and hovered on the popped-up windows to understand the segments within the pie chart, to note them down. It would be easier if there was a way to label the pie chart segments globally, rather than use the approach I followed!   Cheers Roby Double-clicked on over 60 input profiles and hovered on the popped-up windows to understand the segments within the pie chart. Any better approach to globally label the pie chart segments? ... View more

Thank You Melodie, that was very helpful. ... View more