The Design And Inplementation Of The 12366 Big Data Analysis And Mining System Of Shandong Province State Tax Bureau

12366 is a special taxation service line for taxpayers' tax-related issues,providing taxpayers with clear-standard and service-standardized tax law promotion and taxation consulting services.It is a bridge for taxpayers to communicate with the state tax authorities,which is the most direct and efficient way for taxpayers to solve tax issues.On June 3,2011,12366 Taxation Service line was formally settled in the State Taxation Bureau of Shandong Province.It offers tax-related services to taxpayers in Shandong Province with "listen sincerely and serve wholeheartedly”as its guideline.The service line system saves dialogues between taxpayers and service personnel simultaneously as voice files.By the end of 2017,there are 8T voice files that are valid with the format of V3.The voice file storage directory is classified and stored according to the phone call date and seat number of service personnel,including 10 categories of information of consultations,inquiries,complaints,reports of violations of laws and disciplines,tax reports,suggestions,appointments,subscriptions,service transfers,and other issues.Each category can have up to 7 subcategories,with 1,251 in total.The audio files contain a large number of taxpayers' real perceptions and business experiences,reflecting the problems encountered by taxpayers in taxation in a certain period of time.This thesis focuses on the analysis of the status quo and major problems of the 12366 tax service line in China,converting the massive voice data by the tax service personnel into text.Through Natural Language Processing,the voice text is classified and visually analyzed with central issues of taxpayers extracted.By analyzing the needs of the taxpayers,this thesis proposes that appropriate strategies and measures be adopted,the quality of tax services be improved,accurate services to taxpayers be provided,the awareness of tax services for taxpayers be raised,and the existing tax system be supplemented.The system is mainly divided into three major categories of fifteen modules.The three major categories include general business,big data visualization analysis,and big data exploration.The general business includes six modules:system management,basic settings,report management,detailed voice classification,complaint process,and data analysis.Big data analysis includes home page display,regional classification statistics,detailed classification statistics,service personnel working statistics,call duration statistics,and keyword query statistics.Data exploration includes phone number forecasting,hot words analysis and correlation analysis.This system is implemented by B/S architecture,MVC mode,JAVA language with version 1.8.0 25 64 bit,and Eclipse.The middleware adopts WebLogic Server 12c(12.2.1)and database is Oracle 1 1g R2 64-bit.