Clustering Multivariate Empiric Characteristic Functions for Multi-Class SVM Classification

Resum

A dissimilarity measure between the empiric characteristic functions of the subsamples associated to the different classes in a multivariate data set is proposed. This measure can be efficiently computed, and it depends on all the cases of each class. It may be used to find groups of similar classes, which could be joined for further analysis, or it could be employed to perform an agglomerative hierarchical cluster analysis of the set of classes. The final tree can serve to build a family of binary classification models, offering an alternative approach to the multi-class SVM problem. We have tested this dendrogram based SVM approach with the one-against-one SVM approach over four publicly available data sets, three of them being microarray data. Both performances have been found equivalent, but the first solution requires a smaller number of binary SVM models.