Visualization of the Kernel Hierarchical Clustering Using Discriminant Analysis
Hideyuki Haruyama, Yasunori Endo · Medical Entomology and Zoology · 2005
Kernel agglomerative hierarchical clustering (K-AHC) is a technique of clustering in which the data is mapped from the pattern space to a feature space by some kernel function. K-AHC can obtain a good result, but it is unknown how the data of pattern space is mapped from the above thing in the feature space, because of the nonlinearity of the kernel functions. It is very important to visualize a distribution of the data in the feature space for selecting the kernel functions. In this paper, we will discuss the visualization of the data in the feature space using Kernel Fisher Discriminant Analysis(K-FDA).