I have developed two models in unsupervised clustering analysis with K-means(proc fastclus) and hierarchical clustering(proc cluster). Does anyone know to access these two models and tell me which one is better that another? Should I use R squares or other measurements? Thanks you.