Description: The last step in training phase is refinement of the clustersfound above. Although DynamicClustering counters all thebasic k-means disadvantages, setting the intra-cluster similarityr may require experimentation. Also, a cluster mayhave a lot in common with another, ie, sequences assignedto it are as close to it as they are to another cluster. Theremay also be denser sub-clusters within the larger ones.
File list (Check if you may need any files):
Download