Study On The Application For Active Incremental Data Mining In The On-Line Analytical Mining Model

The concept of work warehouse is used. It aims at the problem of complexity of the basic data and consummates the process of basic data on the basis of the classic on-line analytical mining model. The data warehouse and the other kinds of data files are expressed as a work warehouse in which the basic logic structure is data cube. A meta data module and a knowledge base module are added to the work warehouse which makes the data mining on the work warehouse more flexible than on the data files such as data warehouse directly. The function is more abundant and the system integrating degree is higher.The advantages of the on-line analytical processing and the data mining are combined in the on-line analytical mining system. It needs to process the system events more systematic, more efficient and more intelligent after the combination. The incremental data mining function is implemented in order to making the whole system facing the decision users considering the system's entirety, initiative and efficiency. The active incremental mining function is presented that is to process the incremental data automatically when incremental data satisfy the certain condition and trigger the incremental data mining program. That makes incremental data mining function of data mining engine in the system automation. The auto mechanism in an active database is used for reference. The data foundation is made as center in the current relation database and the multi-dimension data warehouse. The system solves the whole incremental mining function of the on-line analytical mining system from the data foundation. The flexibility, intelligent level and running and maintaining efficiency of the system is improved.In order to implement the on-line analytical mining model, the on-line analytical processing and the active incremental mining need to be implemented in the database. The on-line analytical processing system processes the multi-dimension analysis in the virtual multi-dimension data set based on the data warehouse technology of SQL Server 2000. Active incremental data mining is implemented by applying the trigger mechanism of the database to invoke procedures.