Software

Summary: ClusterSenses is a graphical exploration tool for
soft-assigned clustering methods with special emphasis on finite
mixture models. It is entirely based on the concept of brushing; the
ability to select data in one view and have other views reflect this
selection. In ClusterSenses, the cluster assignment is represented by
histograms while the data points are put on a scatter plot. The user
selects a data subset according to common cluster attributes by
clicking on histogram bars and also directly through the data view
brushing individual points. Cluster models and data are supplied by
the user and may be combined rather freely.

details and screenshots at:
http://www.inf.tu-dresden.de/~dk17/cluster-senses

greetings,
daniel kottow
PhD student
CS-department
Dresden Germany