基于(α,λ)联系度容差关系的变精度粗糙集模型

摘要

Traditional rough set-based methods cannot deal with incomplete information system with noisy data effectively. According to set pair analysis, (α,λ) connection degree tolerance relation is defined. This paper introduces the variable precision (α,λ) connection degree rough set model which combines the (α,λ) connection degree tolerance relation with the majority inclusion relation proposed by Ziarko. A heuristic attribute reduction algorithm based on the positive region similarity is given. Time complexity of the algorithm is analyzed. The experimental results show that the proposed method is more effective for incomplete information system with noisy data.%基于传统相糙集理论的方法不能有效地处理含噪音的不完备信息系统.根据集对分析理论,提出(α,λ)联系度容差关系.将(α,λ)联系度容差关系与Ziarko提出的多数包含关系相结合,提出变精度(α,λ)联系度粗糙集模型.给出了该模型下基于正域相似度的启发式属性约简算法,分析了算法的时间复杂度,通过仿真实验验证了所提方法处理含噪音的不完备信息系统的有效性.