Extended Rough Set Model Based on Prior Probability and Valued Tolerance Relation

摘要

Based on equivalence relation, the classical rough set theory is unable to deal with incomplete information systems. In this case, an extended rough set model based on valued tolerance relation and prior probability obtained from incomplete information systems is firstly presented in this paper. As a part of the model, the corresponding discernibility matrix and an attribute reduction of incomplete information system are then proposed. Finally, the extended rough set model and the proposed attribute reduction algorithm are verified under an incomplete information system. Example analysis shows the improved tolerance relation model and the proposed attribute reduction algorithm can deal with incomplete information system effectively.