Introduction

The goal of the Hierarchical Cluster Analysis (HCA) method is to find groupings in the input data matrix $ \mathbf{X}$. The input metric is used to measure the similarity between the objects in the input matrix $ \mathbf{X}$. The distance matrix $ \mathbf{D}$ is a matrix whose elements say something about the similarities or dissimilarities between objects. That is, $ d_{ik}$ says something about the similarity/dissimilarity between object $ i$ and object $ k$.

We start with a warning: Since the number of ways to permute the dataset is $ n!$, and the number of choices of metrics and methods is large, the different methods often give different answers. The risk getting the result ``you want'', which could be wrong, is therefore high. Thus the results should be analysed with great caution.



Bjørn Kåre Alsberg 2006-04-06