基于粗糙集理論的不完備信息系統(tǒng)下的知識獲取方法研究.pdf

1、重慶大學碩士學位論文基于粗糙集理論的不完備信息系統(tǒng)下的知識獲取方法研究姓名：白俊卿申請學位級別：碩士專業(yè)：應用數(shù)學指導教師：劉瓊蓀20060501重慶大學碩士學位論文 英文摘要IIABSTRACTHuman's knowledge is continuously enriched and updated. However, by contrast tothe objective world, it is incomplete,

2、undependable and uncertain. Human beings havebeen continuously and gradually learning the objective world by using the impreciseand incomplete knowledge. Rough set theory is a mathematical tool used for dealingwith vague

3、ness and uncertainty. Simultaneously, the classical rough set theory isbased on complete information. In this paper, Knowledge Discovery based on roughset theory under incomplete information is researched.In the paper, a

4、n overview of the current situation of researches on Rough Set, andthe main issues related to the incomplete data problem and the commonly- used methods of handling incomplete data problems are detailed. Then rough set t

5、heory’ sbasic concepts and properties are summarized. Binary relation and attribute reductionalgorithms of incomplete information system are also summarized. Based on thetheory, probability discernibility matrix is defin

6、ed and corresponding discernibilityfunction is given. Then a probability algorism for attributes reduction is proposed. The probability that the attribute belong to the reduction can be know from thereduction which is ga

7、ined by the algorism, and we can sample according to thisreduction.Along with the rapid increase of data, the incremental data mining has raised wideconcerns. In this paper, attribute price (P) is defined and a new incre

8、mental algorithmfor attribute reduction is provided. The attribute reduction of new information systemcan be got by the algorithm in the dynamic mode when a condition attribute is addedto the information system without t

9、he change of object and decision attribute. At last,the dynamic reduction algorithm is improved and the improved algorithm is given.Keywords: Rough Sets Theory, Incomplete Information System, AttributeReduction, Upper Ap