Research On Data Fusion Technology In Tax Risk Monitoring Platform For Application

In the era of big data,there were more than 40 main business systems of the Local Taxation Bureau of Guizhou Province,generating a large amount of tax-related data.According to the"Risk Management Strategic Plan" of the State Administration of Taxation,with the purpose of effectively using the tax information of various business systems to strengthen the tax risk management of the province,Guizhou local tax revenue risk monitoring platform emerged as the times require.The tax risk monitoring platform incorporates a large number of taxpayer basic information and tax-related data for more than 10 systems.Data fusion among different systems has become a core technical issue in the development of tax risk monitoring platforms.The specific performance is as follows:first is the issue of standardization and integration of cross-platform heterogeneous data,secondly,the taxpayer tax-related information feature points are extracted and categorized and dealt with.The third is the joint judgment of taxpayers with multiple feature risk points.In order to solve the above three main problems,three data fusion methods are adopted:data layer fusion,feature layer fusion,and decision layer fusion.(1)The data layer fusion combines ORACLE database-related technologies,data collection,data collection,and taxpayer basic information and tax-related information for each platform based on the National Golden Tax 3 system,including data reduction and data matching.Cross-platform data fusion;(2)Feature layer fusion According to the actual business requirements,the data in the data layer is processed according to different characteristics.The data will be extracted through data extraction,data cleaning,data conversion,data matching,data aggregation,and data aggregation.The taxation information of the taxpayers in Guizhou Province is categorized and the characteristics of the business domain are used to complete the construction of thematic data warehouse.On this basis,a data mart is set up for the areas of the business departments of the local taxation bureau;(3)The main function of the decision-making layer fusion is to compare the data of taxpayers in data warehouses and data marts with multi-level and multi-dimensional joint judgments through multiple risk characteristic indicators and mathematical models,and to compare the result values of the taxpayers with the early warning values.The final judgment results are obtained,and finally the taxpayer's overall risk information is finally obtained by the method of weighted summation.Off,the process comprising pre-risk risk indicators for each data,feature extraction,recognition judgment,and finally to obtain Integrated Risk Information taxpayer.The tax risk monitoring platform served more than 1,890 local tax bureaus and more than 900,000 taxpayers in the Local Taxation Bureau of Guizhou Province.Through the use of data fusion technology,a total of 354 data tables for more than 10 systems were merged,totaling about 2 billion data pieces.And on a monthly basis,about 900,000 taxpayers across the province identify and determine the characteristics of risk suspects.The system has been put on the line so far,pushing taxpayer risk tasks for 30 issues,identifying 83,751,539 risks,pushing 79,947 taxpayers,and assessing the local taxation bureau of Guizhou Province.The new tax revenue is 6.879 billion yuan,which has achieved good social and economic benefits.