Cluster Analysis Recognition of Inherent Data Structures

Cluster Analysis Recognition of Inherent Data Structures 

Inhomogeneities in data can be studied by cluster analysis. By means of cluster analysis both structures of objects and variables can be found without any pre-information on type and number of groupings (unsupervised learning, unsupervised pattern recognition). 

Geometrically illustrated, clusters are continuous regions of a highdimensional space, each of them containing a relatively high density of points (e.g., objects), separated from each other by regions that are relatively empty (low density of points). The belonging of points (objects) to 

Methods of cluster analysis may be distinguished into two groups  

There exist several methods of hierarchical clustering which use diverse measures of distance or similarity, respectively, e.g., single linkage, complete linkage, average linkage, centroid linkage, and Ward s method (Sharaf et al. [1986], Massart et al. [1988], Otto [1998] Danzer et al. [2001]). 

---