Generalized Rough Set Model Based on Set Pair Connection Degree

An improved classification relation between objects based on existing generalized rough set models and set pair connection degree of set pair analysis theory, called generalized set pair similarity relation is proposed Based on this a more generalized rough set model is presented. UCI database is selected to demonstrate that the classification performance of the new model is superior to those existing models. Therefore the proposed method is more suitable for processing large-scale incomplete information system.