THE FRAMEWORK BASED ON DECISION SUPPORT SYSTEM FOR KNOWLEDGE ACQUISITION FROM IMPORT AND EXPORT DATA

摘要

As P.R. China is affiliated into the WTO, millions of merchandises have been imported and exported. All these trade entries are registered and stored into the Relational Database by leveraging the traditional On-Line Transaction Process. However, how to analyze these massive data to support the government's right decision-making? To this purpose, this paper provides a systematic model based on our practice and implementation. First of all, a star schema (MSDN 2003) with a fact table is constructed. This schema is a container of data so that these OLTP data would be analyzed more efficiently. In the second step, with the help of On-Line Analytical Process, data in the Relational Database is rearranged and then imported into the star schema container to build the data warehouse. In the third step, according to some algorithms such as classification, cluster analysis and so on, knowledge will be discovered by drill-down and roll-up operations on the data warehouse. In the fourth step, discovered knowledge will be stored into the knowledge library for decision support. In the end, a man-machine interface is presented to display the result.