Rough Set Research on Rule Extraction in Information Table

The problem of extracting rules from information table has been studied by use of rough set theory. According to the corresponding relationship and the indiscernible relationship, the concepts of classification quality, discernible vector. Through the discernible vector, we can scan the discernible and obtain the core attribute set and the importance of every attribute. A reduced attribute set can be acquired through combining the score attribute set and the important attribute selected from the different attribute set in the discernible vector with the constraints of classification quality. Finally, delete redundant attribute value and obtain the concise rules. The illustration and experiment results indicate that the method is effective and efficient for rule extraction.