The Research About Attribute Reduction In The Incomplete Information Systems Based On Roughsets

In 1982, Polish scholar Z. Pawlak put forward Rough Set theory. It can be utilizedas a mathematical tool for the analysis of imprecise and incomplete information with the support of the interrelated data set only. So far plentiful achievements of RoughSet have been made both in theory research and application.Data Mining is a process that can abstract embedded and potentially usefulinformation from large amounts of data that are incomplete, noisy, fuzzy, and random.Rough Set can deal with the data containing incomplete information, which is beyond the ability of the traditional technology of data mining. As the extension of set theory, one of Rough Set's central research fields is data mining with incomplete information.In real life, there are a lot of difficuties about data error, date understanding or the data access restrictions, and other factors,so we ofen face the incomplete information system during the acquisition of knowledge, which greatly restricted the rough sets of theory to practical direction. In the international and domestic, rough sets theoretical studies is also unusual. Therefore, the information system related to data processing has become one of the main elements of rough sets research.The dissertation mainly focuses on the attribute reduction of Rough Set in the incomplete information system.[11],[12],[13],[14],[15] obtained some achievements to the incomplete information system research. this design has studied already the incomplete information system correlation research results which has, traced the research about the incomplete information attribute reduction in the domestic and foreign, improved the algorithm of the attribute reduction in the incomplete information system. Proposed own algorithm and have confirmed the algorithm validity.The main work includes:* Proposed the artificial neural networks packing algorithmExtracts the entire information system, resolves it the sub- information system, trains the artificial neural networks using the integrity subsystem. But after based on in the nerve network foundation which trains, completes the incomplete information system to result in the attribute value to pad, we confirm the instance with the empirical datum. * Proposed the arribute redution algorithm based on the attribute importionIn the attribute important foundation, we analyzed the predecessor research resultsand proved this algorithm is valid.