2011 | OriginalPaper | Buchkapitel
Generalized Parameterized Approximations
verfasst von : Jerzy W. Grzymała-Busse
Erschienen in: Rough Sets and Knowledge Technology
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
We study generalized parameterized approximations, defined using both rough set theory and probability theory. The main objective is to study, for a given subset of the universe
U
, all such parameterized approximations, i.e., for all parameter values. For an approximation space (
U
,
R
), where
R
is an equivalence relation, there is only one type of such parameterized approximations. For an approximation space (
U
,
R
), where
R
is an arbitrary binary relation, three types of parameterized approximations are introduced in this paper: singleton, subset and concept. We show that the number of parameterized approximations of given type is not greater than the cardinality of
U
. Additionally, we show that singleton parameterized approximations are not useful for data mining, since such approximations, in general, are not even locally definable.