Name : [in Japanese]
Number : 17
Location : [in Japanese]
Date : November 13, 2004 - November 14, 2004
Pages 153-156
A cluster analysis method is proposed in this paper. This method is able to visualize a multi-dimensional dataset as a graphical object, or glyph, and extract a tree structure from the object as a dendrogram. As a benchmark, the fisher's iris dataset was used. As a result, the dendrogram took 17.3 percent of error, but the glyptic visualization helped to locate these errors. To obtain more useful dendrogram which has the corresponding visualization, use of this method is recommended.