Data Classification Based On Rough Set Knowledge Discovery Method And Its Application

With the information age of high-speed development, database storage capacity has also become explosive growth. In order to find important information from the screening contains rules, meet the needs of development, data classification technology has become an important direction of information processing. This paper briefly introduced the classification; this paper focuses on the research of classification method based on rough set theory.Rough set effective data analysis technology, without any prior knowledge of the case, to dig out the important data or rules, and the rules are easy to understand, operation form based on the rough.Set theory based on redefining the attribute importance concept; focus on the attribute reduction in two aspects. Attribute reduction is maintained in the knowledge classification ability invariable to delete redundant attributes, based on the definition of absolute nucleus of the algorithm as the foundation, while maintaining the positive domain invariable situation, by comparing the attributes important degree to improve the algorithm, this algorithm can reduce computation quantity, increase the efficiency of computation. In the classification rules, redefine the nuclear values, combining the attribute importance degree, are given for inconsistent decision table reduction method of classification rules.In this paper,based on rough set theory classification method is applied to ship fuel.Consumption data analysis,simulation experiment shows that the algorithm is effective.