Probabilistic decision tables in the variable precision rough set model

The Variable Precision Rough Set Model (VPRS) is an extension of the origin al rough set model. This extension is directed towards deriving decision ta ble-based predictive models from data with parametrically adjustable degree s of accuracy. The imprecise nature of such models leads to quite significa nt modification of the classical notion of decision table. This is accompli shed by introducing the idea of approximation region-based, or probabilisti c decision table which is a tabular specification of three, in general unce rtain, disjunctive decision rules corresponding to rough approximation regi ons: positive, boundary and negative regions. The focus of the paper is on the extraction of such decision tables from data, their relationship to con junctive rules and probabilistic assessment of decision confidence with suc h rules.