I don't understand this solution. What is the sense of clustering on a single target variable? I thought the whole idea of clustering was to find clusters of observations that are "similar" and then perhaps might related to a target variable. How is clustering on a single target variable any different from looking at a distribution of that variable and seeing if there are any groups? And, from your solution, what are the X,Y coordinates of the constellation (when there is a single variable being clustered)?