论文部分内容阅读
文章针对Rough Set理论的核心内容之一属性约简进行了研究。结合信息论的有关知识，研究了在属性约简过程中决策属性集相对条件属性集的条件熵的变化规律，在此基础上提出了新的属性约简算法。实验分析表明，在多数情况下这种算法都能够得到决策表的最小约简，同时还对算法复杂度做了简单的分析。
In this paper, attribute reduction is one of the core contents of Rough Set theory. Combining with the relevant knowledge of information theory, this paper studies the changing rule of conditional entropy of the relative set of condition attributes in the attribute reduction process, and proposes a new attribute reduction algorithm based on this. Experimental analysis shows that in most cases, this algorithm can get the minimum reduction of decision table, and at the same time makes a simple analysis of the complexity of the algorithm.