Geology Reference
In-Depth Information
Fig. 3.3 Different types of clusters illustrated as 2D points [
72
]. a Well-separated clusters Each
point is closer to all of the points in its cluster than to any point in another cluster, b Center-based
clusters Each point is closer to the center of any other cluster, c Contiguity-based cluster Each
point is closer to at least one point in its cluster than to any point in another cluster, d Density-
based clusters Clusters are regions of high density separated by regions of low density,
e Conceptual clusters Points in a cluster share some general property that derives from the entire
set of points (Points in the intersection of the circles belong to both)
3.4.1 Hierarchical Tree Cluster Analysis
Hierarchical CA is comprised of agglomerative methods and divisive methods
which
find clusters of observations within a data set. Hierarchical clustering clas-
sifies data by means of a series of partitions which may run from one single cluster
of all elements to
n
clusters, each containing one element. The main constituent of
the analysis is the repeated calculation of distances between objects or clusters. The
graphical outcome of this approach is known as a dendrogram. Divisive methods
and agglomerative methods are two major groups of methods for hierarchical CA,
in which the agglomerative methods are popular in the research community.
Commonly used algorithms in hierarchical clustering are single linkage clustering,
complete linkage clustering, average linkage clustering, average group linkage, and
Ward
s linkage. Each of these methods differs in the way that similarity or distance
between an element and a group of elements is de
'
ned, and consequently produces
different results using the same data which is detailed in Fig.
3.4
.
