基于区间二型模糊粗糙集的连续属性约简算法

摘要

一型模糊粗糙集可以直接处理连续属性集，但不能处理高度不确定性数据，而区间二型模糊集可以增强系统处理不确定性的能力。为了提高处理噪声数据的精确度，在一型模糊粗糙集的基础上，定义区间二型模糊粗糙集。基于区间二型模糊粗糙集模型研究了连续域决策信息系统的属性约简，通过紧计算域给出了新的约简算法。由于拒绝变量集合的存在，提出的约简算法可在有限时间内收敛，并且得到了更加合理的结果。数值仿真验证了约简算法的可行性与有效性。%Type-1 fuzzy rough sets could directly handle datasets with real-valued attributes and couldn’t effectively deal with highly uncertain data.Interval type-2 fuzzy sets enhance the system’s ability to handle uncertainties.To improve the accuracy in processing noise data,it defined interval type-2 fuzzy rough sets on the basis of type-1 fuzzy rough sets.It discussed the at-tribute reduction of continuous domain decision information system,and then given a reduction algorithm based on compact computational domain.Because of the existence of the refused variable set,the proposed reduction algorithm could be conver-gence in a limited time and obtain more reasonable results.Finally,it demonstrated the feasibility and effectiveness of the re-duction algorithm by simulations.