基于粗集理论的数据预处理及应用研究-计算机系统结构专业论文.docxVIP

重庆大学硕士学位论文 重庆大学硕士学位论文 英文摘要 ABSTRACT With the rapid development of databases techniques and computer network,large amount ofdata is stored,the rapid growth demand for extracting，understanding and assimilating useful knowledge from the growing mountains of data outspaces the traditional methods of data analysis，which leads to the emerging field of knowledge discovery in databases and data mining．Rough set theory is new maths tools，and its characteristic have no need of other existing information，which make it overcome shortcoming of other methods and avoid the influence of subjective factor to the results ofdatamining．and become olle ofprimary methods ofKDD． Because data preprocessing have an important influence on KDD and Rough set， Solving these problems efficiently can improve efficiency,exactitude and availability of pattern in application ofrough set． In this thesis，we study and discuss deeply data preprocessing based on rough set． First，the characteristic and shortcoming of primary algorithms about the completation ofnull values is analysised，and principle and target ofthe completafion is indicated．According to the shortcoming of data completafion algorithms based on rough set，a strategy for data completation based on valued similarity relation，and a strategy based on the limited similarity relation are put forward to improve effect of completation．Moreover,a strategy for imputation of null values based on changed precision model is put forward to improve anti—disturbing ability ofmodel Secondly,several primary algorithms are indroueed and discussed，the direction and target of discretization is analysised，To gain logical cuts，a method to discretize continuous attributes based on Rough Entropy is brought forward At last，the potential client datamining system in electronic commerce is put forward，All given algorithms is applied to the system，and compared by test r