Abstract: The information view of rough set theory is discussed in the light of an extended information view. A core attributes computation algorithm of decision table based on extend information entropy is proposed. And a from-bottom-to-top decision table attribute deduction algorithm which takes the significance of the attribute as the heuristic information is designed. Meanwhile, the including value of attributes is adopted as the second standard to choose attribute in order to make the reduction faster. The new algorithm “EIEAAR” can deal with both the consistent and inconsistent decision tables, and integrate the core attribute computation and non-core attribute reduction in a whole. At last, the complexity of the algorithm is analyzed and two kinds of examples are taken to test the validity of the algorithm. The experiment shows that the algorithm is valid.

摘要： 从一种扩展的信息观的角度出发，讨论了Rough集理论的信息论观点。提出了一种基于扩展的信息熵的决策表核属性计算算法，并设计了基于属性重要性为启发信息的自下而上的决策表属性约简算法EIEAAR。同时针对不一致表，将属性对不相容对象的包含值作为第二标准选择属性以加快约简速度。EIEAAR算法能处理一致和不一致决策表，并将核属性计算和非核属性约简统一起来。最后，对算法进行复杂度分析并用实例验证算法的有效性。实验表明该算法能有效得到决策表的最小约简。