【摘要】 基于容差关系和先验概率容差关系的粗糙集模型是粗糙集理论的重要扩充,但却均有其局限性。本研究结合上述两种模型的优点,提出了一种改进的先验概率粗糙集模型,新模型是建立在对不完备决策系统属性值统计数据的基础上,既考虑了同一属性取值的不同情况,又考虑到不同属性之间的关联性,可以有效提高分类精度和分类的合理性。该模型对属性间存在明显的关联性且未知属性值较少的系统具有很强的实用性。更多还原

【Abstract】 Rough set models based on tolerance relation and a prior probability tolerance relation are important expansions of the rough set theory, yet there are some limitations. Combining the advantages of the two models, an improved prior probability rough set model is proposed. The new model is based on the statistics of property values in incomplete decision system, taking into account both the same attribute’s different values and the correlation between different attributes. The model can effectively improve classification accuracy and rationality. The model has a strong practicability in system which has fewer unknown attribute values and exists a significant association in different attributes.更多还原