Design And Implementation Of Tax Risk Administration System Based On Data Mining

The system for tax risk administration is objectively required by the development of the taxation system, by the demand of tax collection and administration, by the development of information technology as well.The system in this thesis was based on the OECD risk model, and used mainly J2 EE and other mature technologies, and integrated data mining model into its design and development.This system broke the entire process of risk administration up into four links: management of risk characteristics, risk identification and assessment, risk response and feedback. It identified risks, abstracted risk characteristics, scanned generated risk information, sorted and assessed the result. Then tasks of risk were handed down as a whole through the risk arranger. After tasks were fulfilled, we could check out, inspect and handle. Finally we got feedback and evaluation on accuracy and rationality of risk characteristics.This thesis set forth design and implementation of this system in detail. In architecture design, this system was decomposed into risk management platform and task management platform. This thesis demonstrated every detail, such as analysis of business requirement, design of business process and function module, implementation and test for this system. The thesis presented all interfaces of system function. We integrated data mining model into this system, and took land value added tax as an example to explain how to use indexes for mining features in order for seek risk points.This thesis analyzed development requirement, system architecture, and then expounded design and implementation of this system. We studied the theory of tax risk, and integrated the data mining model, and then transformed the theory into the software by this way. At the same time, the system was more scalable, portable and flexible when decomposed into risk management platform and task management platform in architecture design.