IFPD confirmation measures are used in ranking inductive rules in Data Mining. Many measures of this kind have been defined in literature. We show how some of them are related to each other via weighted means. The special structure of IFPD measures allows to define also new monotonicity and symmetry properties which appear quite natural in such context. We also suggest a way to measure the degree of symmetry of IFPD confirmation measures.
Monotonicity and Symmetry of IFPD Bayesian Confirmation Measures
CELOTTO, EMILIO;ELLERO, Andrea;FERRETTI, Paola
2016-01-01
Abstract
IFPD confirmation measures are used in ranking inductive rules in Data Mining. Many measures of this kind have been defined in literature. We show how some of them are related to each other via weighted means. The special structure of IFPD measures allows to define also new monotonicity and symmetry properties which appear quite natural in such context. We also suggest a way to measure the degree of symmetry of IFPD confirmation measures.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
Monotonicity and Symmetry of IFPD Bayesian Confirmation Measures.pdf
non disponibili
Tipologia:
Documento in Post-print
Licenza:
Accesso chiuso-personale
Dimensione
291.62 kB
Formato
Adobe PDF
|
291.62 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
