Proactive Data Mining System And Application In The Freight Forwarding Business

Data mining is a process of extracting valid, previously unknown, comprehensible, and usable information from large databases. Along with the development of information technology, more and more data from various fields are produced extraordinary. At the same time, data mining technology is being developed continually and transferred to application gradually. Along with the rapid development of freight company, it should be a trend that data mining be applied to freight company's data analysis.Active data mining is a new direction of data mining. To obtain more expert andauthentic result, user-centered active data mining can supply various methods which apply the domain knowledge and related experience from domain expert in KDD procedure.At the beginning of this paper, describe the content of the active data mining, explain the history of freight company, analyze the recent research of the active data mining and the data mining application in the freight forwarding business. Then the theory research in the active data mining and the application scope of data mining in freight company are introduced.Base on the research of the active data mining technique and freight forwarding business, design and implement a user-centered active data mining system. The system model is a user-centered active data mining model, supplies three system units in mining procedure (active data collection, user-centered active data mining, active user feedback unit) to user which can be repeated in a screw method. The system structure is a multilayer system structure with five layers. They are user interactive layer, system program interface layer, mining core service layer, data access layer, database layer. The system frame is implemented with the MVC pattern. The subsystems which have high independence and extension, such as data mining arithmetic subsystem, mining result visualization subsystem, are implemented with the plug-in system base on XML technique. The implementation of each subsystem is introduced in detail.