Apart from connectionist approaches and genetic algorithms, for the most part the methods of inductive concept learning share the common objectives of classifying and producing predictive knowledge from observations. Although the rules produced are generally required to be intelligible and accurate, some problems arise due both to the complexity of the description languages and the noise and uncertainty in the initial observations.
Incorporating statistical techniques into empirical symbolic learning systems
ESPOSITO, Floriana;MALERBA, Donato;SEMERARO, Giovanni
1993-01-01
Abstract
Apart from connectionist approaches and genetic algorithms, for the most part the methods of inductive concept learning share the common objectives of classifying and producing predictive knowledge from observations. Although the rules produced are generally required to be intelligible and accurate, some problems arise due both to the complexity of the description languages and the noise and uncertainty in the initial observations.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.