Decision rules induced from a data set allow to particularize the relationships between condition and decision factors. Several indices can be used to characterize the most significant decision rules based on "historical" data, but they are not able to measure the impact that these rules (or strategies derived from these rules) will produce in the future. Thus, in this paper, a new methodology is introduced to quantify the impact that a strategy derived from decision rules may have on a real life situation in the future. The utility of this approach is illustrated by an example.
Measuring the expected impact of decision rule application
Greco, Salvatore
;Matarazzo, Benedetto;Pappalardo, Nello;
2004-01-01
Abstract
Decision rules induced from a data set allow to particularize the relationships between condition and decision factors. Several indices can be used to characterize the most significant decision rules based on "historical" data, but they are not able to measure the impact that these rules (or strategies derived from these rules) will produce in the future. Thus, in this paper, a new methodology is introduced to quantify the impact that a strategy derived from decision rules may have on a real life situation in the future. The utility of this approach is illustrated by an example.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.