Mining Classification Rules Base On Rough Set Theory

With development of database technology and the coming of in formation era, largeamount of data is accumulating in many industries. The volumes of the database arezooming.In order to improve the efficiency of work and quality of life, people have toderive valuable knowledge embedded in data from databases. For the aim, people havebegun there search on knowledge discovery in databases. As we all know, however,usually there are redundant data, missing data, uncertain data and inconsistent data in thedatabases and they become a great barrier to extracting knowledge from databasesRough Sets(RS) theory was put forward by Pawlak Zdzislaw in 1982. After morethan 20 Years of developing, it has received fruitful achievements in both of theory andapplications. RS doesn't depend on additional information beyond the data which is apotent tool for dealing with imprecise, incomplete vage and uncertain data. Sometradditional method of knowledge discovery is only suitable for precise set not for roughset. Since many set of data in real life is rough, the model of knowledge discovery basedon Rough Sets Theory plays an importent role in information system.Firstly, the history, status and possible development direction of KDD are in troducedand the main methods and techniques of KDD are also reviewed. Secondly, the roughsets theory is introduced and general application procedure of rough sets theory in KDDis analyzed. In the paper, an overview of the cunent situation of researches on Rough Set,and the main issues related to the incomplete data problem and the commonly-usedmethods of handling incomplete data problems are detailed. On research results ofpredecessors basis, a comprehensive method is proposed, which can simultaneouslyderive rules from incomplete data sets based on rough sets. In addition, rough sets anddecision tree have complementary characteristics. A new approach to generation adecision_tree based on rough sets is thus proposed combining both advantages. Theexperiment results show that these methods advanced and practica.