论文部分内容阅读
属性约简是粗糙集理论中的核心问题之一,概念格是进行知识表示和数据分析的一种有效工具.文中利用概念格作为约简工具,给出基于概念格的多层属性约简算法,提出相融可辨概念、相融等价概念、亏n级等概念,研究内涵亏值对分类能力变化产生的影响,给出概念格中形式背景约简的判定定理.文中算法能完备地求出所有可约简的最大属性集合,从而为概念格中属性约简提供一种有效方法.最后,通过实例分析和实验对比说明该约简算法的可行性与有效性.
Attribute reduction is one of the core problems in rough set theory, concept lattice is an effective tool for knowledge representation and data analysis.With the concept lattice as a reduction tool, a concept-based multi-layer attribute reduction algorithm , Proposed concepts such as the concept of blending and distinguishing, the concept of blending equivalence and the concept of deficit and so on, and studied the influence of the connotation value on the change of classification ability, and provided the judgment theorem of formal context reduction in concept lattice. Find all the reducible maximum attribute sets, so as to provide an effective method for the attribute reduction in the concept lattice.Finally, the feasibility and validity of the reduction algorithm are illustrated through the example analysis and experimental comparison.