Some silhouette-based graphics for clustering interpretation

Some silhouette-based graphics for clustering interpretation

0.00 Avg rating0 Votes
Article ID: iaor19901158
Country: Belgium
Volume: 29
Issue: 3
Start Page Number: 35
End Page Number: 55
Publication Date: Sep 1989
Journal: Belgian Journal of Operations Research, Statistics and Computer Science
Authors: , ,
Keywords: cluster analysis
Abstract:

Silhouettes were developed as a graphical display for nonhierarchical cluster analysis. They are based on the ratio between the tightness of a cluster and its separation from other clusters. A possible extension is to represent for each object both these characteristics in a two dimensional graph. The same technique can also be used with fuzzy clustering, making use directly of the fuzzy membership functions to measure the tightness of the links of each object with its principal cluster and its neighbour.

Reviews

Required fields are marked *. Your email address will not be published.