Re:The lsm command for Latent Semantic Mapping
by microtherion on Friday June 24, 2011 @04:48PM
(#2693806)
You're right that agglomerative clustering (unlike K-means) does not inherently need to know the # of clusters in advance. However, it still needs some sort of termination criterium, and # of clusters is one possible criterium.
Since lsm operates in a transformed space, other commonly used criteria like cluster distance may not be as convenient for the user to express.
What I've found is that the problem seems to get a lot more reasonable if you know how many clusters there are.
K-Means requires this information, but afaict agglomerative techniques don't. I wonder why this tool's agglomerative clustering method requires the number of clusters as an argument.