基于粗集的不完备信息系统属性约简

Rough Set-based Attribute Reduction under Incomplete Information Systems

属性约简是粗集理论研究的核心内容之一。经典粗集理论是建立在完备信息基础之上的 ,然而在现实中 ,不完备信息系统的广泛存在极大地限制了粗集理论向实用化迈进。该文基于相容关系 ,将分布约简、最大分布约简、分配约简引入不完备信息系统 ,提出了一种新的约简———分配序约简 ,并讨论了几种约简之间的关系。给出了分配约简的一种启发式算法 :条件信息量约简算法 ,分析了该算法的时间复杂度。经实验检验 ,该算法是有效的。

Knowledge acquisition based on rough set theory is an important and difficult task under incomplete information systems. Several reduction methods like distribution reduction, maximum distribution reduction, and assignment reduction are introduced into incomplete information systems, and a new reduction,namely,assignment order reduction,is defined. The relationship among them is examined as well. Information quantity and conditional information quantity are defined to express indispensable attributes under incomplete information systems. Based on conditional information quantity, a heuristic algorithm for assignment reduction is presented, and the complexity of this algorithm is analyzed. Finally, the experimental result shows this algorithm can find its assignment reduction for incomplete information system.