Clustering with a distance matrix

Question

I have a (symmetric) matrix M that represents the distance between each pair of nodes. For example,

    A   B   C   D   E   F   G   H   I   J   K   L         


        

Answer 1

Hierarchical clustering works directly with the distance matrix instead of the actual observations. If you know the number of clusters, you will already know your stopping criterion (stop when there are k clusters). The main trick here will be to choose an appropriate linkage method. Also, this paper(pdf) gives an excellent overview of all kinds of clustering methods.