【摘要】 求核和属性约简是粗糙集理论研究的一个核心问题。文中主要针对现有的一些决策表属性约简算法存在的不足,尤其是基于信息熵的属性约简算法在较大数据集上效率不高的问题提出改进。主要通过结合粗糙集的相关理论来改进原有的属性约简算法在求核中的约束条件,进而在原有算法的基础上提出了一种改进算法。在求约简属性集时,利用新提出的约简算法,使计算复杂度降低,同时保持了高效的决策准确率。实验结果表明改进后的决策表属性约简方法能够更加快速有效地找到约简集。更多还原

【Abstract】 Searching core and attribute reduction is a main issue of the rough sets theory.To solve some existing shortcomings of the decision table attribute reduction algorithm,in particular,entropy-based algorithm has low efficiency for reduction of large data sets,so it proposed an improved algorithm based on the theory of rough sets.The new algorithm changed the constraint condition in searching core through using some rough sets theory.It has high efficiency and has low time complexity in searching core and attribute reduction.Experiment results show that the algorithm can find a good attribute subset.更多还原