Methodology of rough-set-based classification and sorting with hierarchical structure of attributes and criteria

摘要: We consider a hierarchical classification problem involving sets of attributes and criteria. The concerns an assignment set objects to pre-defined classes. preference-ordered classes is called sorting. are described by two sorts attributes: criteria regular attributes, depending on whether the attribute domain or not. sorting made in finite number steps due structure form tree. propose methodology based decision rule preference model. model constructed inductive learning from examples decisions Decision Maker reference objects. To deal with inconsistencies appearing we adapt rough approach problems. Due inconsistency their propagation bottom top hierarchy, description object particular may be not simple value but either subset interval criterion scale. An example illustrates presented.