Determining the Number of Clusters/Segments in Hierarchical Clustering/Segmentation Algorithms
We investigate techniques to automatically determine the number of clusters to return from hierarchical clustering and segmentation algorithms. We propose an efficient algorithm, the L Method, that finds the "knee" in a '# of clusters vs. clustering evaluation metric' graph. Using the knee is well-known but is not a particularly well-understood method to determine the number of clusters. We explore the feasibility of this method, and attempt to determine in which situations it will and will not work.
Salvador, S., Chan, P.K. (2003). Determining the number of clusters/segments in hierarchical clustering/segmentation algorithms (CS-2003-18). Melbourne, FL. Florida Institute of Technology.