Research on information system attribute set information granules based on functional dependency

Granular computing is a useful method in the computational intelligence field. The current research on information granule is mainly granulation of information system object set and its properties. The concept of information granules of information system attribute set based on functional dependency is proposed, and then the concepts of information system structure and information system structure complexity are defined. The changing rules of information system structure are studied when function dependency or attribute changes. A new concept of attribute reduction is presented and efficient calculation method is given.