Cluster Dissimilarity
In order to decide which clusters should be combined (for agglomerative), or where a cluster should be split (for divisive), a measure of dissimilarity between sets of observations is required. In most methods of hierarchical clustering, this is achieved by use of an appropriate metric (a measure of distance between pairs of observations), and a linkage criterion which specifies the dissimilarity of sets as a function of the pairwise distances of observations in the sets.
Read more about this topic: Hierarchical Clustering
Famous quotes containing the word cluster:
“Next week Reagan will probably announce that American scientists have discovered that the entire U.S. agricultural surplus can be compacted into a giant tomato one thousand miles across, which will be suspended above the Kremlin from a cluster of U.S. satellites flying in geosynchronous orbit. At the first sign of trouble the satellites will drop the tomato on the Kremlin, drowning the fractious Muscovites in ketchup.”
—Alexander Cockburn (b. 1941)