A NEW FEATURE-SELECTION METHOD TO EXTRACT FUNCTIONAL STRUCTURES FROM MULTIDIMENSIONAL SYMBOLIC DATA

In this paper, we propose a feature selection method to extract functi onal structures embedded in multidimensional data. In our approach, we do not approximate functional structures directly. Instead, we focus on the seemingly trivial property that functional structures are geome trically thin in an informative subspace. Using this property, we can exclude irrelevant features to describe functional structures. As a re sult, we can use conventional identification methods, which use only i nformative features, to accurately identify functional structures. In this paper, we define Geometrical Thickness (GT) in the Cartesian Syst em Model (CSM), a mathematical model that can manipulate symbolic data . Additionally, we define Total Geometrical Thickness (TGT) which expr esses geometrical structures in data. Using TGT, we investigate a new feature selection method and show its capabilities by applying it to t wo sets of artificial and one set of real data.