ClusteringDemo (anyone have a better name?) is a demonstration of different clustering algorithms.
It is written entirely in JavaTM and can be run as a
standalone application or as an applet (probably without load/save capabilities, depending on permissions).
The applet can be tried here,
though you might see an older version. You can download the latest version from the
project page.
Version history
2.1.0
pending
Improved user interface and caching of intermediate results. Tree display of clusters.
2.0.2
October 29, 2003
Small improvements.
To do
Online help, tooltips and documentation
Information on each algorithm
More algorithms
Different quality-of-partition measurements
Localization
...
Screenshots
Standard view
Points are displayed graphically. Black points are unassigned. Otherwise, all points of the same
colour belong to the same cluster. For the clusterer used in this example, points are also connected
by lines to make it easier to see any changes.
Tree view
Clusters and points are displayed as nodes in a tree. Points give their coordinates, clusters give
their compactness as a fraction.
