Application Research Of Data Mining Techniques In Tax Risk Management System

With the development of Internet technology,people's working life cannot be separated from the Internet.According to the strategy Internet+Taxation proposed by the State Administration of Taxation,the Henan Provincial Taxation Bureau deployed online application,online reporting,taxation risk management and other systems on the basis of the official launch of the Golden Tax III,which facilitates taxpayers' declaration and payment.Taxation staff's levy management.In order to generate some valuable forecasting results for tax officials to prevent risks,this paper analyzes various data mining techniques and uses association rule methods for tax risk forecasting.This paper first analyzes the background of tax risk management,sums up various data mining technologies,and designs the Apriori algorithm and FP-Growth algorithm for discovering association rules according to the demand of tax risk forecasting.Experiments show that the latter has better time efficiency.Then with the tax data of the whole 2017 year of city Zhengzhou as a sample,the FP-Growth algorithm to collect the frequent set of the tax-deficit companies,analyze the associated rules,and generate tax risk forecast results.The results show that the association rules method is effective.Finally,data mining based on association rules is implemented as a module in the tax risk management system.