Automatic extraction of clusters from hierarchical clustering representations

摘要: Hierarchical clustering algorithms are typically more effective in detecting the true structure of a data set than partitioning algorithms. However, hierarchical do not actually create clusters, but compute only representation set. This makes them unsuitable as an automatic pre-processing step for other that operate on detected clusters. is both dendrograms and reachability plots, which have been proposed representations, different advantages disadvantages. In this paper we first investigate relation between plots introduce methods to convert into each showing they essentially contain same information. Based then technique automatically determines significant clusters cluster representation. it time possible use requires no user interaction select from