Parameterized imprecise classification: Elicitation and assessment

Abstract

This work is based on classifiers that can yield possibilistic valuations as output. The valuations may have been obtained from a labeled data set either directly as such, by possibilistic classifiers, by transforming the output of probabilistic classifiers or else by adapting prototype-based classifiers in general. Imprecise classifications are elicited from the possibilistic valuations by varying a parameter that makes the overall classification become more or less precise. We introduce some indices to assess the accuracy of the parameterized imprecise classifications and their reliability, thus allowing the user to choose the most suitable level of imprecision and/or uncertainty for a given application. © Springer-Verlag Berlin Heidelberg 2006.