A method of rough set based on the attribute oriented induction and relevance analysis

For certain database with huge data, such as marketing database, rule mining and knowledge acquisition could be a difficult problem. Aimed at this point, a new method is proposed. On the basis of data preprocessing, it uses the attribute-oriented induction and the attribute relevance analysis to generalize customer attributes and eliminate redundant attributes in these databases and to constitute a contracted decision-making information system. And then extracting valuable decision-making rules with the help of reduction method of decision rules in rough set theory. It has been proved to be with higher efficiency and great practical significance by the following example.