In order to reduce the hybrid decision systems including the missing value attributes, which are lost or represent “do not care” conditions, a generalized incomplete rough set model based on neighborhood relations, the discrimination methods of the missing value and a hybrid reduction algorithm were proposed. The model approximates an arbitrary subset in the universe with neighborhood granules, and the generalized neighborhood relations are the generalization of the asymmetry similarity relations and the tolerance relations. The discrimination methods of the lost or “do not care” conditions were proposed based on the assumption of the consistency classification, and the influence of the noise samples and the neighborhood values to the classification accuracy was presented as well. The validity and feasibility of the algorithm were demonstrated by the results of experiments on five UCI machine learning databases.
