Knowledge acquisition in incomplete information systems based on variable precision rough set model

This paper deals with knowledge acquisition in incomplete information systems using variable rough set model. We introduce the concepts of β-lower and β-upper approximations. We also propose reduction of knowledge that eliminates only that information, which is not essential from the point of view of classification or decision making within β precision. In our approach we make only one assumption about unknown values: the real value of a missing attribute is one from the attribute domain. We show how to find decision rules directly from such an incomplete decision table, which are as little non-deterministic as possible and have minimal number of conditions.