Research and implement of classification rule mining algorithm based on attribute reduction

This paper brings up a new classification algorithm of data mining (CRMA) in any scale relation database. Based on Rough set theory it divides relation table into several equivalence class based on attribute values, calculates information capacity in decision factor of the every condition attribution, eliminates redundancy attributions, and erases repeat units. Then classification rules can be obtained through strong equivalence class which relation table was reduced. It overcomes the redundancy nature, complicated nature and unfit nature to big capacity data or increment data of some classification algorithm at present. It has higher efficiency and widespread application perspective in large and incremental databases. The mining algorithm and an example are discussed in details.