Clustering with Variable Importance - Filtering and K-Selection
I have been using the Tibco clustering with variable importance data function which has the benefit of limiting the number of clusters to a statistically significant number, but I would still like to have the option of using this limit or not. I am a novice with R Scripting and cannot figure out how to modify the function to allow me to do this, so I would like to know if any other users have created a modified version exist that allows the user to select the Hartigan K or user input K to complete the analysis?
Also - when selecting the input variable to be limited by the active filter or markings I find that it randomly selects rows and I end up with a mix of rows that are within and outside of my filters that are used in the cluster analysis. Does anyone have a fix for this?