0

I have analyzed the following Data set in R using hierarchical analysis, and plotted the results. Ive been asked to find and highlight the number of clusters present, but how exactly do you determine this? It seems it all depends where you cut the plot at?

For example, with saying there are 5 clusters, you get something like this: enter image description here

But with saying there are 2 clusters, you get something like this: enter image description here

What is the right cluster amount, and how do you determine this?

FEAR
  • 1
  • The number of cluster is part of the output from the cutree() function. It is easier to demonstrate if you can provide a sample of your data and code. – Dave2e Apr 02 '22 at 12:38
  • 2
    The NbClust package will compute 30 different indices indicating the "right" number of clusters. In many situations, the "right" number of clusters is related to what you are planning to do with the clusters. There is no guarantee that there is a unique "right" number of clusters for all purposes. – dcarlson Apr 02 '22 at 13:53

0 Answers0