贪心约简算法在HSV中的应用

本文研究了粗糙集理论中的属性约简问题。一般的约简算法和改进的约简算法都不能够得到一个令人满意的属性约简结果。为了找到具有较少属性的约简,文中提出了使用贪心约简算法,通过对接受过超选择性迷走神经切断术(HSV)治疗的具有11个属性的20个十二指肠溃疡病人构成的信息系统作近似分析,获取了一个与原决策表分类质量相同的仅含有5个属性的较小属性集。实验证明:用此方法能有效地去除冗余信息,对其症状进行约简提炼,从而获取简单而又能体现症状与病征的规则。

An Application of Greedy Reduction Algorithm in HSV

In this paper,the author studies the attribute reduction of rough set theory.The attribute reduction extracted from ordinary reduction algorithm and improved reduction algorithm is not satisfactory.The author proposes that the method of greedy reduction algorithm is used to approximate the analysis of an information system describing 20 patients with duodenal ulcer treated by HSV(highly selective vagotomy).The patients are described by 11 attributes.At last the less sets of attributes ensuring an acceptable quality of classification are calculated,which has been reduced to 5 relevant attributes.It is proved in the experiment that in this way,superfluous knowledge can be removed easily,symptoms can be reduced and a corresponding rule can be extracted.