Next:Multiresolutional
Cluster DisplayUp:Hierarchical
Parallel CoordinatesPrevious:Hierarchical
Clustering
Visualizing Clusters
Each node T_{i} in a hierarchical cluster tree T
represents a nested collection of enclosed data points or subclusters.
The node contains a summary of all points and subclusters rooted from
it. The following information may be directly obtained from T_{i}.

n_{i} : the number of data points enclosed.

m_{i} : the mean of the data points.

B_{i} : the extents, i.e. the minimum and maximum bounds
of the cluster for each dimension.

v_{i} : a measure of the size of cluster T_{i}.

l_{i} : the tree depth at node T_{i}
v_{i} is a computed measure of a cluster size and satisfies
the following criteria: If T_{i} is an ancestor of T_{j},
then
v_{i}
> v_{j}
The value of v_{i} is directly dependent on the shape
of the clusters produced by the clustering algorithm. For spherical clusters,
v_{i}
may be the radius of a cluster. For rectangular clusters,
v_{i}
may be the
Ndimensional volume of the cluster.
In parallel coordinates, we display the information at a node by making
use of variablewidth opacity bands. Figure 2
shows a graduated band that visually encodes the information for a cluster.
The mean stretches across the middle of the band and is encoded with the
deepest opacity. The deepest opacity is a function of the density of a
cluster, defined as the ratio .
The top and bottom edges of the band have full transparency. The opacity
across the rest of the band is linearly interpolated. The thickness of
the band across each axis section represents the extents of the cluster
in that dimension.
Figure 2:A single multidimensional graduated
band that visually encodes information at a cluster node.

Next:Multiresolutional
Cluster DisplayUp:Hierarchical
Parallel CoordinatesPrevious:Hierarchical
Clustering