概念层次中基于粗糙集的优化可信规则获取.pdfVIP

第 31 卷第 2 期 2011 年 2 月 计算机应用 VOI. 31 NO.2 Feb.2011 JOumal Of COmputer Applicati Ons 文章编号 :1∞1 -9081 (2011 )02 -0493 -05 dOi: lO. 3724/SP. J.1087. 201 1.∞493 概念层次中基于粗糙集的优化可信规则获取 梁德翠，胡培 (西南交通大学经济管理学院，成都 61∞31 ) (decuiliang@ 126. com) 摘 要:随着系统中数据量剧增，规则太多以及不同决策者对规则有不同层次需求等问题，概念层次提供了一种 解决方法讨论条件属性具有概念层次的情况下，利用粗糙集理论分析属性在不同层次组合下的正浅和规则关系， 自顶向下提出了概念层次中基于粗糙集的优化可信规则获取的算法。该算法改进了现有的属性约简策咯，借助描述 子实现属性约简并获取优化可信规则。考虑到层次上正域为空和正域没有新增对象的特殊情况，提高了规则获取的 效率。最后通过实例分析说明该算法的可行性。 关键词:概念层次;描述子:属性约简;优化可信规则;粗糙集 中图分类号: TPI8 文献标志码 :A Optimal credible rules based on rough se臼 in concept hierarchies LIANG De-cui, HU Pei (School 01 Ecorwmics aru1 M.α阳ge阳1ú， Southwest Jiaotong University , Chengdu Sichuan 61∞31 ， China) Abstract: The cOncept hierarchies provide a new way tO solve such prOblems as the data are increasing t∞ fast， there 缸e 回 many rules, and different decisi On makers need Obtain sOme rules 仕om different hierarchies. In the c Ontext that every c Onditi On attribute had a cOncept hierarchy, the relati Ons Of positive regiOns and rules from different levels were analyzed based On rough set theo巧. Then, the algOrithm Of Optimal credible rules based On rOugh sets in cOncept hierarchies w.嗣 proposed， which was alOng with cOncept hierarchies frOm top tO dOwn. The algOrithm imprOved the existing reducti On strategy, realized the reducti On and Obtained the Optimal credible rules by descriptOr百. Besides, it alsO cOnsider穹d there was null Or nOt a new Object in the positive regi On, and the effectiveness Of extracting rules was improved. Finally, the feasibility of this algOrithm was validated by an example. Key words: cOncept hierarchies; descriptOr; attribute reducti On; Optimal credible rule; rough set 0 引言 数据挖掘是从大量原始数据(信息)中发现知识的过程， 而知识常以规则形式呈现。但在实际获取规则过程中，存在 以下问题: 1 )系统中数据量巨大，因而获取的规则太多[1] ; 2)建立在原始数据水平下，发掘出的模式常常难以理解，往 往需耍更简单和更综合的规则 21 ;3) 在利用规则决策过程 中，不 [r?J类型的决策者对参考的规则会有不同层次水平的要 求，这也就安求能提供不同层次的规则。另外，在实际决策过 程中，获取精确数据需要花费很大成本其歪很难获取。 概念层次L 1-4 J 为以 k问题提供了一种解决思路。在面向 属性归纳方法中，概念层次被定义成