Decision Trees for Three-Way Data

Abstract: The framework of this short paper is tree-based models for three-way data. Three-way data are data that are classified in three ways. Three way data are obtained when prior information play a role in the analysis. Thus, they can be derived when a stratifying variable is used to distinguish either groups of variables or groups of objects. Three way data can be analysed by exploratory methods as well as confirmatory methods. Recently, we have introduced a methodology for classification and regression trees in order to deal specifically with three-way data. So far we have developed only partitioning procedures discarding the induction point of view