Tax Risk Management Plan
一、Background
1、Information Technology Development Background
In recent years, information technology has developed rapidly, especially in the past two years. The vigorous development of cloud computing and big data technology, the core technology from concept to theory, from experimental environment to successful implementation, in turn, has promoted the further development of technology. With the continuous expansion of network bandwidth and the rapid development of mobile Internet, information application systems have rapidly expanded from computer terminals to tablet and mobile terminals. The rapid development of information technology combined with mobile Internet has promoted the informatization development of all walks of life and put forward higher requirements for the informatization and modernization of government departments. In particular, the State Council of the PRC issued the Guiding Opinions on Actively Promoting ‘Internet+’ Actions, which opened the door to the ‘Internet +’ era in China and put forward higher requirements for the informatization of government departments.
For a long time, the tax system has actively used modern information technology to transform and improve the core content of tax management services and has made remarkable achievements. Since the 1990s, the informatization of tax collection and management has gradually started. From the first, second and third phases of the golden tax, efforts have been made to put the tax work on the network and support the complex tax business system with advanced information systems. Jun Wang, director of the State Administration of Taxation, pointed out pointedly that without tax informatization, there would be no tax modernization, and put forward the goal of basically realizing tax modernization by 2020. In the process of informatization of the tax system, it is necessary to ‘be good at taking advantage of all situations and opportunities’ and use ‘Internet’ technology to form a ‘Internet + tax’ plan and action plan. In terms of tax services, it is necessary to develop an electronic tax bureau to realize that taxpayers do not go out of their homes to handle taxes. In tax collection and management, we strengthened the application of new technologies and established a big data platform for data analysis.
2、Internet+
After the national ‘Internet +’ strategy was put forward, various departments and industries have launched ‘Internet + medical’, ‘Internet + logistics’ and so on. All walks of life are actively exploring the transformation and upgrading under the new normal. More and more data are sent, disseminated, and stored through the Internet to reflect the basic information, business information and investment information of enterprises. There is more and more information on the Internet, accounting for more and more, there are more and more types. Because the Internet data covers enterprise inventory management, purchase and sales, financial accounting, etc., the data is true and reliable. By comparing with the taxpayer declaration data within the tax authorities, the authenticity and accuracy of enterprise declaration can be directly reflected. The method is simple and effective. Of course, there are many other types of data on the Internet, which cannot be stored in the traditional binary relationship table. In the past, this type of data can only be discarded. However, from practical experience, this kind of information can well print the actual business situation of enterprises, such as signed contracts, personnel recruitment, etc. Driven by the rapid development of the market economy, the business mode of enterprises is becoming more and more complex. Relatively speaking, the management mode of the tax department continues the traditional way. Through the simple comparison of data and based on personal experience, an index system is designed to reveal the possible risks of the enterprise. In the face of doubts, the
information is not effective enough and the analysis process is not comprehensive enough.
Under the planning and guidance of the State Administration of Taxation, local tax systems have actively carried out the action plan of ‘Internet + taxation’, and Suzhou State Administration of Taxation of Jiangsu Province has built an online tax service system with ‘Smart Suzhou’ as the starting point. Shenzhen Municipal State Administration of Taxation adheres to the ‘pain point thinking’ and speeds up the construction of ‘Internet + electronic tax bureau’, etc. From the practice of various places, in the field of tax service, it is better to explore and try by using the Internet thinking, that is, to transplant the business handled by the entity tax office to the Internet. Through the Internet, it can be handled with computers, tablets, mobile phones and other terminals, to facilitate taxpayers to handle tax. At the same time, through the push of information on various service platforms and the interactive exchange between tax enterprises, taxpayers can obtain relevant information through various channels to provide services for taxpayers.
3、Tax Risk Management Analysis
In terms of tax management, the Internet + thinking mode is rarely mentioned, especially in terms of tax risk management. The main reasons are as follows:
(1) The risk analysis is based on the taxpayer declaration data and the internal data of the tax department, and rarely extends to other data sources.
(2) Risk analysis is mainly based on structured data, rarely involving unstructured data.
(3) Due to the lack of platform support, the analysis of data mainly focuses on simple comparison, and less involves the establishment of data model and the implementation of data mining algorithm.
Based on the above analysis, in the ‘Internet +’ era, there is an urgent need to build a platform to automatically obtain all kinds of information on the Internet, including e-commerce data, bidding data, industry website data, land market network, group purchase network, listed company information, recruitment information, business contracts, credit ratings, etc., through general collection and non-fixed-point collection procedures, and obtain third-party information from government departments through interfaces, And the internal data of the tax department, through integration and screening, form information that reflects the overall picture of the enterprise, and provide help for the management of the tax department, including risk analysis and risk response.
二、Introduction to Big Data Application Platform
2.1 Construction Objectives
The construction goal of big data application platform is to establish an integrated tax related data application platform, complete data collection, integration, analysis, and other functions, and provide support for the management of tax departments.
The main features of the platform include:
- Automation
The data collection, screening, integration, and analysis are automatically completed by the platform according to the pre-set.
- Integration
The platform includes data processing, analysis, and response functions.
- Intellectualization
Based on Hadoop big data architecture platform, use data mining algorithm to realize value-added utilization of data.
- Convenience
The user can obtain relevant information and get the desired results by clicking the mouse.
The big data application platform realizes massive Internet data collection with a wider range of data types. It is not only limited to structured data, but also includes semi-structured and unstructured data. For semi-structured data, it can be stored after structured through semantic analysis. For unstructured data, it can be automatically classified through application models. Realize data integration, use data mining algorithms to mine valuable data from massive Internet information, and form a panoramic display for taxpayers through data cleaning, processing and conversion, and collection according to individual taxpayers.
There are two methods to collect Internet data:
One is general information collection, which collects relevant data from the Internet by setting the collection path and website and saves them in the database after semantic analysis.
One is non directional collection, which obtains the top ranked information by setting keywords such as taxpayer name, saves the web page information, and establishes a word segmentation index for relevant keywords, to improve the efficiency of web page retrieval. Realize the panoramic display of data, not only display forms and documents, but also view pictures, listen to audio, play video and other types of files.
The platform provides an interface through which data can be received from other application systems.
2.2 Function Positioning
(1) Data Collection
Function Positioning
(1) Data Collection
General information collection: collect relevant data regularly through the item setting menu, and save the collected data to the database after structuring through semantic analysis. The general information collection content can be maintained by the users themselves, and the collection content can be increased or reduced as needed. At present, it mainly includes the announcement information of Shanghai Stock Exchange and Shenzhen Stock Exchange, securities investment fund information, bidding website winning information, Taobao store information, Taobao auction information, T-mall store transaction information, land auction information, etc. Non-fixed-point collection is mainly used to obtain valuable information of a single enterprise on the Internet. Through the search engine, it can obtain enterprise related information, such as address, telephone and other basic information, investment, contract signing and other business information. Two problems need to be solved for non-point collection: one is to obtain information, and the other is to filter the obtained information through algorithm design, filter the worthless information and obtain valuable data.
(2) Data Integration Processing
After sorting and processing the collected Internet information, the structured data is directly stored in the warehouse, and the unstructured data is classified and stored on the data platform. The data platform based on the underlying Hadoop big data architecture can integrate all kinds of data. In general, it is divided into Internet information, third-party data, internal data of tax departments and internal data of enterprises. The data can be structured or unstructured. Based on integrating all kinds of data, the data platform supports analysts to use data, analyze and model the data, and find risk enterprises. At the same time, it uses big data technology to carry out correlation analysis through the implementation of data mining algorithms, to make the data sound, maximize the analysis efficiency and improve the hit rate of risk objects.
(3) Presentation of Data
The data platform should facilitate the retrieval and display of various types of data, including unstructured data such as pictures, audio, video, and traditional structured data. In addition, it does not need to install client programs. It only needs a browser to browse files and analyze data. For the data in the data platform, it provides the functions of user-based collection and panoramic display. It is convenient to analyze a single enterprise. The data range includes Internet information, third-party data, internal data of the tax department and internal data of the enterprise, reflecting the basic information of the enterprise, business information, investment information, declaration information, tax information, etc., forming a complete and focused portrait of the taxpayer.
2.3 Object of Use
Need Further Refinement
This platform is intended for tax department managers, including risk analysts and risk response personnel.
(1) Risk analysts, through the platform’s Internet information collection function and subsequent data integration, processing and analysis functions, use the platform’s data comparison, model building and other tools, from macro comparison to meso trend, to micro observation, and general industry analogy to find risk enterprises from a large number of taxpayers, and can provide doubt reports to support risk response personnel.
(2) After receiving the enterprise response task, the risk response personnel can conduct a comprehensive physical examination of the enterprise through the doubtful point report provided by the data platform and the powerful single household analysis function, simply and clearly grasp the enterprise information, analyze the existing doubtful points, point out the existing risks, conveniently and quickly complete the desk analysis, and can put the clues found in the response process, the collected data, and the sorted materials, Upload to the data platform to reflect the whole response process, response results, enterprise risk elimination and enrich enterprise information.
三、Function Menu
Function Menu
- System Settings
It is used for some system settings required for project operation, including project configuration, permission management, etc.
(1) Project Settings
Configure the fixed-point collection items. Through the configuration, you can increase the collection of structured Internet information. As shown in the following figure.
(2) Project sub item settings, used to set the settings of related sub items in each project. For example, in large and small non projects, related shareholder information, issuance information, etc., set links, collection cycle, whether to collect in full, etc. As shown in the following figure.
(3) Collection Field Settings:
It is used to structure information and store it in database tables to prepare for data utilization. As shown in the following figure:
- Task Management
It is used for the management of acquisition tasks, including the setting of acquisition time and cycle. Log information such as the duration of the collection, the size of the collected data, and whether an error occurred shall be recorded.
- Data Matching
Match the data collected through the Internet with the in-flight taxpayers to prepare for the analysis and utilization of the data. This includes automatic matching, manual matching, and establishing matching rules.
Compare the data collected on the Internet with the registration information of the tax department through keywords such as taxpayer name, tax number, industrial and commercial registration certificate number, and automatically match the information that can be matched and establish matching rules. For those that cannot be accurately matched, the fuzzy matching function is provided. The operator can select the matching object, establish the matching rules and save the matching rules. In the future, the data will be automatically matched according to the established matching rules, and there is no need for manual intervention.
- Data Management
Through the data management function, data collected from the Internet, tax related data of government departments, internal data of tax departments and other data are collected by household. Through data exchange, the data collected on the Internet can be integrated and compared with the internal data of the tax department. At the same time, an interface is reserved to facilitate access to tax related data of government departments and other types of data.
- Data Analysis
Through the data management function, the platform integrates four types of data, including Internet collected data, tax related data of government departments, internal data of tax departments and other data, not only limited to structured data, but also unstructured data. On this basis, risk analysts can find risk enterprises through data integration, self-setting of risk indicators, and comparison of internal and external data. Risk response personnel can query all tax related information of the response enterprise, including structured data and all kinds of unstructured data, and can browse directly without installing plug-ins through the one-user function.
First, implement simple comparison, such as comparison between e-commerce data and internal declaration data
Second, carefully select existing indicators to form a system.