Shaded similarity matrix is a visualization approach used
in hierarchical cluster analysis. This paper takes the unified
point on viewing classification and clustering (supervised
and unsupervised learning) from some researchers [22, 18,
8]. This leads us to believe shaded similarity matrix can
also be used in the task of classification visualization.
In this paper, we investigated how shaded similarity matrix
is used to visualize two popular classification methods:
nearest neighbor and decision tree. We also explored a
method to solve large data sets by using ensemble classification
and sampling techniques.