Tax risk assessment report generation method and device, equipment and medium

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a tax risk assessment report generation method and device, equipment and a medium, and the method comprises the steps: constructing a system, and generating a system data table containing dimension information and index information; performing formula definition, and calculating index data of the associated system data table through the formula definition; performing tax risk definition to generate a risk rule table; determining abnormal system index data according to the risk rule table and submitting and auditing the abnormal system index data, and generating a risk list after the auditing is passed; and generating a tax risk report based on the risk rule table, the risk list and the system data table. By constructing a system, generating a system data table and performing formula definition to calculate index data, tax-related data can be systematically integrated and processed, and the accuracy of risk assessment is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and particularly relates to a method, device, equipment and medium for generating a tax risk assessment report. Background Art

[0002] In order to conduct tax management and risk prevention and control work more scientifically and efficiently, enterprises need to regularly issue tax risk reports to ensure their tax compliance.

[0003] The traditional way of issuing tax risk reports mainly relies on enterprise finance and tax-related personnel. It is necessary to consult and analyze a large amount of business data, carefully verify and confirm abnormal data, and finally summarize and organize it by professional personnel. This way has many drawbacks: on the one hand, issuing risk reports involves multi-person division of labor and cooperation, and the labor cost and time cost are extremely high. Enterprises need to invest a large amount of human and time resources to complete this work; on the other hand, with the continuous expansion of the enterprise scale, the types and quantities of business are becoming increasingly complex. Relying solely on manual identification of tax risks will inevitably lead to omissions and it is difficult to comprehensively and accurately identify potential risks, resulting in tax compliance risks for enterprises.

[0004] In terms of the current situation of industry technology development, although some enterprises and institutions have tried to use some technical means to assist tax risk assessment, the existing solutions still have deficiencies. Most of the existing technologies often cannot efficiently integrate the tax risk-related data scattered in the enterprise during the data processing and risk identification processes, and it is difficult to deeply mine the potential risks behind the data, resulting in a significant discount in the accuracy of risk assessment and unable to meet the growing tax management needs of enterprises. Summary of the Invention

[0005] Aiming at the problem that most of the existing technologies often cannot efficiently integrate the tax risk-related data scattered in the enterprise during the data processing and risk identification processes, resulting in a significant discount in the accuracy of risk assessment and unable to meet the growing tax management needs of enterprises, the present invention provides a method, device, equipment and medium for generating a tax risk assessment report.

[0006] In the first aspect, the technical solution of the present invention provides a method for generating a tax risk assessment report, including the following steps: S1: Construct a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; S2: Select any system, calculate the indicator values under each combination according to the Cartesian product combination of the dimension values under this system, and fill the dimension value combinations and the indicator values into the initial system data table to obtain a system data table; S3: Confirm the combination of dimension values that need risk assessment, set tax risks for the indicators that need risk assessment, and generate a risk rule table; wherein, the risk rule table includes an abnormal judgment rule and the combination of dimension values applicable to this rule. S4: Determine the abnormal indicator data in the system data table according to the abnormal judgment rule in the risk rule table, submit it to the review system, and generate a risk list after passing the review. S5: Generate a tax risk report based on the risk rule table, the risk list, and the system data table.

[0007] As an optimization of the technical solution of the present invention, before step S1, it includes: Define dimension information and indicator information. The dimension information includes a dimension number, a dimension name, and a dimension data source. The indicator information includes an indicator number, an indicator name, and an indicator accuracy.

[0008] As an optimization of the technical solution of the present invention, step S2 includes: S21: Select any system, calculate the Cartesian product of each dimension value in the system to generate all combinations of dimension values. S22: For each indicator in the system, determine the dimensions and calculation formulas on which the indicator calculation depends. S23: Traverse all combinations of dimension values, substitute the dimension values in each combination into the calculation formula of the corresponding indicator, and calculate the indicator value of each indicator under this combination. S24: Fill the initial system data table with each combination of dimension values and the calculated indicator values to obtain the system data table.

[0009] As an optimization of the technical solution of the present invention, step S22 includes: S221: Take an indicator in the system as the output value, and all dimension information in the system or several other indicators in the same dimension system as the input values. S222: Use the input values to perform custom mathematical operations to construct the calculation formula of the output indicator. S223: Load the calculation formula of the output indicator, analyze all its dependent indicators; judge the legality of the current formula, that is, judge whether there is a situation of indicator circular dependence. If the formula is legal, execute step S23; if the formula is illegal, execute step S222.

[0010] As an optimization of the technical solution of the present invention, step S223 includes: S2231: Load the calculation formula of the output indicator, analyze all its dependent indicators, and the output indicator is the source indicator. S2232: Judge the dependent indicators in turn. S2233: If there is a situation where the dependent indicator is an indicator data fetching and the dependent indicator is equal to the source indicator, it is determined as circular dependency, and an illegal prompt message is returned. S2234: After all the dependent indicators of the source indicator have completed the parsing loop judgment, if no circular dependency situation is found, the formula corresponding to the source indicator is legal.

[0011] As an optimization of the technical solution of the present invention, step S3 includes: S31: Confirm the dimension combination that needs risk assessment under the selected system. S32: Analyze the index nature and business association of the indicators in the system, obtain the association between each indicator and tax risk, and determine the indicators that need risk assessment. S33: Set tax risks for the indicators that need risk assessment and generate a risk rule table.

[0012] As an optimization of the technical solution of the present invention, step S5 includes: S51: Integrate the exception judgment rules and dimension value combinations in the risk rule table, the exception indicator data in the risk list, and the dimension information and indicator information covered by the system data table. S52: Classify and summarize the exception indicator data in the risk list according to the exception judgment rules and risk types in the risk rule table. S53: Convert and adapt the integrated data and the classified and summarized data into the format required by the large model. S54: Submit the adapted data to the large model through the interface or platform connected to the large model, set the relevant parameters of the large model according to the requirement of generating a tax risk report, and at the same time, input an instruction for generating a tax risk report based on the submitted data to the large model. S55: Obtain the tax risk report generated by the large model after analyzing and processing the submitted data according to its own algorithm and training data after receiving the data and instruction.

[0013] As an optimization of the technical solution of the present invention, the method further includes: When the review fails, re-determine the values of the indicator categories according to the review suggestions in combination with the latest business data and industry standards until the review passes.

[0014] In a second aspect, the technical solution of the present invention further provides a tax risk assessment report generation device, including a system construction module, a formula definition module, a risk definition module, a dimension combination determination module, a risk list generation module, and a risk assessment report generation module; The system construction module is used to construct a system based on a number of dimensions and a number of indicators, and each system generates an initial system data table containing dimension information and indicator information. A formula definition module, which is used to select any system, calculate the respective index values for each combination according to the Cartesian product combinations of the respective dimensional values in the system, and fill the dimensional value combinations and the respective index values into the initial system data table to obtain a system data table; A risk definition module, which is used to confirm the dimensional value combinations that need risk assessment, set tax risks for the indicators that need risk assessment, and generate a risk rule table; wherein, the risk rule table includes an abnormal judgment rule and the dimensional value combinations applicable to this rule; A risk list generation module, which is used to determine the abnormal index data in the system data table according to the abnormal judgment rule in the risk rule table when there is a corresponding dimensional combination in the risk rule table, submit it to the audit system, and generate a risk list after passing the audit; A risk assessment report generation module, which is used to generate a tax risk report based on the risk rule table, the risk list, and the system data table.

[0015] As an optimization of the technical solution of the present invention, the device further includes a dimension and index definition module, which is used to define dimension information including a dimension number, a dimension name, and a dimension data source, and index information including an index number, an index name, and an index accuracy.

[0016] As an optimization of the technical solution of the present invention, the formula definition module includes a dimensional value combination generation unit, a formula definition unit, an index data calculation unit, and a system data table generation unit; The dimensional value combination generation unit is used to select any system and calculate the Cartesian product of the respective dimensional values in the system to generate all dimensional value combinations; The formula definition unit is used to determine the dimensions and calculation formulas on which an index in the system depends for each index in the system; The index data calculation unit is used to traverse all dimensional value combinations, substitute the dimensional values in each combination into the calculation formula of the corresponding index, and calculate the index value of each index under this combination; The system data table generation unit is used to fill the respective dimensional value combinations and the calculated respective index values into the initial system data table to obtain a system data table.

[0017] As an optimization of the technical solution of the present invention, the formula definition unit includes a formula definition sub-module and a legality judgment sub-module; The formula definition sub-module is used to take an index in the system as an output value, and all dimensional information in the system or other several indexes of the same-dimensional system as input values; use the input values to perform custom mathematical operations to construct the calculation formula of the output index; The legality judgment sub-module is used to load the calculation formula of the output indicator, analyze all its dependent indicators, where the output indicator is the source indicator; judge the dependent indicators in sequence; if there is a situation where a dependent indicator is an indicator data fetch and the dependent indicator is equal to the source indicator, it is determined as a circular dependency and an illegal prompt message is returned; when all the dependent indicators of the source indicator have completed the parsing and circular judgment, if no circular dependency is found, the formula corresponding to the source indicator is legal.

[0018] As an optimization of the technical solution of the present invention, the risk definition module is specifically used to confirm the dimension combination that needs risk assessment under the selected system; analyze the index nature and business association of the indicators in the system to obtain the association between each indicator and tax risk, and determine the indicators that need risk assessment; perform tax risk settings on the indicators that need risk assessment and generate a risk rule table. Based on the analysis of the index nature and business association, determine the corresponding category for each indicator in the system; where the category includes positive indicators, inverse indicators, and moderate indicators; save the risk definition and synchronously generate a risk rule table.

[0019] As an optimization of the technical solution of the present invention, the device further includes a risk warning module, which is used to perform risk warning on the indicator data of the system data table according to the risk definition and the risk rule table, automatically submit the warning information for review, and transfer the risk warning information to the risk list after the review is passed.

[0020] As an optimization of the technical solution of the present invention, the risk assessment report generation module includes a collection and collation unit, a format conversion unit, a submission setting unit, and an acquisition unit; The collection and collation unit is used to integrate the exception judgment rules and dimension value combinations in the risk rule table, the exception indicator data in the risk list, and the dimension information and indicator information covered by the system data table; classify and summarize the exception indicator data in the risk list according to the exception judgment rules and risk types in the risk rule table; The format conversion unit is used to convert and adapt the integrated data and the classified and summarized data according to the format required by the large model; The submission setting unit is used to submit the adapted data to the large model through an interface or platform connected to the large model, set the relevant parameters of the large model according to the need to generate a tax risk report, and at the same time, input an instruction to generate a tax risk report based on the submitted data to the large model; The acquisition unit is used to acquire the tax risk report generated by the large model after analyzing and processing the submitted data according to its own algorithm and training data after receiving the data and instruction.

[0021] In a third aspect, the technical solution of the present invention further provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the tax risk assessment report generation method as described in the first aspect.

[0022] In a fourth aspect, the technical solution of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions cause the computer to execute the tax risk assessment report generation method as described in the first aspect.

[0023] As can be seen from the above technical solutions, the present application has the following advantages: By constructing a system and generating system data tables, and performing formula definitions to calculate index data, tax-related data can be systematically integrated and processed. This method changes the situation where manual data processing is prone to errors and low efficiency in the past, improves the accuracy and efficiency of data processing, and provides a reliable data basis for subsequent risk assessment.

[0024] Confirm the dimension combination of risk assessment, determine abnormal system index data according to the risk rule table and submit it for review, and finally generate a risk list, making the whole process standardized. This process optimization avoids the randomness and repetitive work of manual operations, improves the efficiency of risk assessment, and at the same time ensures the reliability and effectiveness of the risk list. When generating a tax risk report with the help of a large model, relevant data is processed to improve the accuracy of risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the present application, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a schematic flowchart of the method provided by the embodiment of the present invention.

[0027] Figure 2 It is a schematic block diagram of the device provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To make the application purpose, features, and advantages of this application more obvious and understandable, the following will use specific embodiments and accompanying drawings to clearly and completely describe the technical solutions protected by this application. Obviously, the embodiments described below are only a part of the embodiments of this application, rather than all embodiments. Based on the embodiments in this patent, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this patent.

[0029] As Figure 1 shown, an embodiment of the present invention provides a method for generating a tax risk assessment report, including the following steps: S1: Construct a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; Before this step, dimension definition and indicator definition are also required; Dimension definition includes dimension number, dimension name, and dimension data source, such as accounting books, accounting organizations, income tax organizations, tax organizations, tax period types, tax periods, etc.

[0030] Indicator definition includes indicator number, indicator name, and precision (indicator values are all numerical types), such as main business income, main business cost, sales expenses, management expenses, financial expenses, input tax amount, output tax amount, etc. The indicators here include source indicators (obtained from business data sources) and derivative indicators (calculated from source indicators and derivative indicators).

[0031] It should be noted that the dimensions and indicators in the system composition are the columns of this data table.

[0032] S2: Select any system, calculate the indicator values for each combination according to the Cartesian product combination of each dimension value under this system, and fill the dimension value combinations and the calculated indicator values into the initial system data table to obtain a system data table; This step specifically includes: S21: Select any system, calculate the Cartesian product of each dimension value within the system to generate all dimension value combinations; the Cartesian product means that in mathematics, the Cartesian product of two sets X and Y is denoted as X × Y, which is the set of all possible ordered pairs (x, y), where x belongs to X and y belongs to Y. In this step, it is to combine the values of each dimension within the system to generate all possible dimension value combinations. Calculate the Cartesian product of each dimension value.

[0033] S22: For each indicator within the system, determine the dimensions and calculation formulas on which the indicator calculation depends; This step specifically includes: S221: Take one indicator in the system as the output value, and take all dimensional information in the system or several other indicators in the same-dimensional system as the input values; For example, in a system that includes indicators such as "main business income" and "main business cost" and dimensions such as "accounting books" and "tax payment period", if "main business profit" is taken as the output indicator, then other indicators such as "main business income" and "main business cost" and related dimensional information can be used as input values.

[0034] S222: Use the input values to perform custom mathematical operations to construct the calculation formula for the output indicator; S223: Load the calculation formula of the output indicator and analyze all its dependent indicators; judge the legality of the current formula, that is, judge whether there is a situation of circular dependence of indicators; If the formula is legal, execute step S23; if the formula is illegal, execute step S222.

[0035] Correspondingly, step S223 includes: S2231: Load the calculation formula of the output indicator, analyze all its dependent indicators, and the output indicator is the source indicator; S2232: Judge the dependent indicators in turn; S2233: If there is a situation where a dependent indicator is an indicator data fetch and the dependent indicator is equal to the source indicator, it is determined as circular dependence, and an illegal prompt message is returned; S2234: When all the dependent indicators of the source indicator have completed the parsing loop judgment, if no circular dependence situation is found, the formula corresponding to the source indicator is legal.

[0036] S23: Traverse all combinations of dimensional values, substitute the dimensional values in each combination into the calculation formula of the corresponding indicator, and calculate the indicator values of each indicator under this combination; S24: Fill each combination of dimensional values and the calculated indicator values into the initial system data table to obtain the system data table.

[0037] After determining a system, take one indicator in the system as the output value, take all dimensional information in the system or several other indicator information (excluding the current indicator) in the same-dimensional system as the input values, and form a calculation method through custom mathematical operations. This process is called defining a formula.

[0038] The system classifies formula types into indicator formulas, sql data fetch formulas, extended logic formulas, etc. Among them, the indicator formula is a formula type that requires four arithmetic operations through other formulas. Different types of formulas play different roles in the data calculation and processing process. For example, the sql data fetch formula is used to obtain specific data from the database and provide a data basis for the calculation of other formulas. If there is the following formula: Index A = Index B ÷ Index C – 1; Index B = Current period sales (retrieved from SQL); Index C = Index D * 0.17 - Index A; These three formulas cannot be calculated in actual calculations. Because when calculating A, the calculation result of C is sought, and the result of C requires Index A. This leads to the concept of index dependence. Index A depends on Index C, and Index C depends on Index A.

[0039] Taking the three formulas of Index A, Index B, and Index C as examples below, the process of circular dependence resolution is illustrated: ① Load the formula of Index A (hereinafter referred to as the source Index A), and analyze all its dependent indexes (which are Index B and Index C).

[0040] ② Loop through and judge Index B and Index C. If Index B is retrieved from SQL, continue to judge the next dependent index C; if Index C is not retrieved from SQL but is an index retrieval, judge whether it is equal to the source Index A. If they are equal, it is determined as a circular dependence, and a prompt message indicating the existence of circular dependence is returned, that is, the formula is illegal. If they are not equal, continue to parse Index C. For example, in the above formula, Index B is retrieved from SQL, continue to judge Index C, Index C is not retrieved from SQL and is not equal to the source Index A, and proceed to the next step.

[0041] (1) Load the formula of Index C, and analyze all its dependent indexes (which are Index D and Index A); (2) Loop through and judge Index D and Index A. If Index D is retrieved from SQL, skip it; whether Index A is retrieved from SQL, obviously not, Index A is an index retrieval, then judge whether A is equal to the source Index A. If they are equal, it is judged as a circular dependence, and a prompt indicating the existence of circular dependence A->C->A is returned.

[0042] ③ After the loop judgment of all the dependent indexes of Index A is completed, Index A is a legal formula and is allowed to be saved.

[0043] The specific process is as follows: Loading the calculation formula of the output index, analyzing all its dependent indexes, and the specific steps to judge the legality of the current formula include: Load the calculation formula of the output index, analyze all its dependent indexes including Index B and Index C, and the output index is the source index; Judge the dependent indexes in turn; If Index B is retrieved from SQL, continue to judge the next dependent index C; If Index C is an index retrieval, judge whether Index C is equal to the source index; If they are equal, it is determined as a circular dependence, and an illegal prompt message is returned; If they are not equal, load the formula for metric C and analyze all its dependent metrics including metric D and metric A; If metric D is retrieved by SQL, skip it and determine whether metric A is equal to the source metric A; If they are equal, it is determined that there is a circular dependency and an illegal prompt message is returned; After all the dependent metrics of metric A have completed the parsing loop judgment, if no circular dependency is found, the formula corresponding to metric A is legal.

[0044] On the basis that the formula definition is completed and the formula is legal, the system calculates using the data in the system according to the defined formula, so as to obtain the metric data of the associated system data table. For example, through the formula for "main business profit" that has been defined and verified as legal, combined with the metric data such as "main business income" and "main business cost", calculate the specific values of "main business profit" under different tax periods, different accounting books and other dimensions, and fill them into the corresponding system data table.

[0045] S3: Confirm the combination of dimension values that need risk assessment, set tax risks for the metrics that need risk assessment and generate a risk rule table; among them, the risk rule table contains abnormal judgment rules and the combination of dimension values applicable to the rules; Specifically include: S31: Confirm the dimension combination that needs risk assessment under the selected system; S32: Analyze the metric nature and business association of the metrics in the system, obtain the association between each metric and tax risk, and determine the metrics that need risk assessment; S33: Set tax risks for the metrics that need risk assessment and generate a risk rule table.

[0046] Before setting categories for each metric in the system, it is necessary to clearly understand the meanings of different categories. A positive metric means that the larger the metric value, the better. For example, the operating income of an enterprise. Generally, the higher the income, the better the operating condition of the enterprise; the opposite is true for a negative metric, where the smaller the metric value, the better. For example, the tax late payment of an enterprise. The smaller the amount, the better the tax compliance of the enterprise; a moderate metric requires the metric value to fluctuate within a specific range. For example, the tax burden rate of an enterprise. Too high or too low may imply tax risks.

[0047] Based on tax business knowledge and the actual operation of the enterprise, analyze the correlation between each indicator and tax risks. Taking "input tax amount" and "output tax amount" as examples, "output tax amount" can generally be classified as a positive indicator. The higher its value, the better the enterprise's sales business and the more taxable income it creates under the premise of legality and compliance. The "input tax amount" needs to be judged in combination with the actual procurement and operation scale of the enterprise. If the procurement activities of the enterprise are normal, an excessive input tax amount may pose a risk of abnormal deduction. In this case, it can be considered as a moderate indicator and should be kept within a reasonable range and matched with the enterprise's operation scale and output tax amount.

[0048] Based on the analysis of the nature of the indicators and business correlations, determine the corresponding categories for each indicator in the system. For example, in a system including indicators such as "main business income", "main business cost", and "sales expenses", "main business income" is set as a positive indicator, "main business cost" is set as an inverse indicator when cost control is reasonable, and for some expense indicators that are greatly affected by industry characteristics and business strategies and have a reasonable range, such as "sales expense ratio", it can be set as a moderate indicator.

[0049] After setting, form a document or enter the corresponding category of each indicator into the system for convenient subsequent reference and verification. At the same time, organize relevant tax professionals and financial personnel to confirm to ensure that the set indicator categories meet the actual tax risk judgment needs of the enterprise and can effectively assist the risk assessment work.

[0050] S4: Determine the abnormal indicator data in the system data table according to the abnormal judgment rules in the risk rule table, submit it to the review system, and generate a risk list after passing the review.

[0051] Judge whether the indicator data is abnormal according to the general rules or principles in the risk definition. For example, if the risk definition stipulates the value range of certain indicators, and the indicator value in the system data table exceeds this range, it is judged as abnormal. According to the rules of the risk definition, comprehensively scan the indicator data in the system data table, identify the abnormal data, and extract it to form a set of abnormal system indicator data.

[0052] S5: Generate a tax risk report based on the risk rule table, risk list, and system data table.

[0053] Specifically include: S51: Integrate the abnormal judgment rules and dimension value combinations in the risk rule table, the abnormal indicator data in the risk list, and the dimension information and indicator information covered by the system data table; S52: Classify and summarize the abnormal indicator data in the risk list according to the abnormal judgment rules and risk types in the risk rule table.

[0054] S53: Convert and adapt the integrated data and the classified and summarized data into the format required by the large model; S54: Submit the adapted data to the large model through an interface or platform connected to the large model, set the relevant parameters of the large model according to the requirements of generating a tax risk report, and at the same time, input an instruction to generate a tax risk report based on the submitted data into the large model; S55: Obtain the tax risk report generated by the large model after analyzing and processing the submitted data according to its own algorithms and training data after receiving the data and instructions.

[0055] It should be noted that this method further includes: When the review fails, re-determine the values of the index categories according to the review suggestions in combination with the latest business data and industry standards until the review passes.

[0056] If it is found in the later review that the index category setting is unreasonable, it needs to be modified in time. First, the reviewer details the reasons for non-passing, for example, the fluctuation characteristics of a certain index in the actual business do not match the set category. Then, the person responsible for risk definition re-analyzes the index, refers to the latest business data and industry standards, and re-determines the appropriate index category. After the modification is completed, submit it for review again until the review passes, ensuring that the index category in the risk definition is accurate and effective and can provide a reliable basis for tax risk judgment.

[0057] As Figure 2 shown, an embodiment of the present invention further provides a tax risk assessment report generation device, including a system construction module, a formula definition module, a risk definition module, a dimension combination determination module, a risk list generation module, and a risk assessment report generation module; The system construction module is used to construct a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; The formula definition module is used to select any system, calculate the respective indicator values of the system according to the Cartesian product combinations of the respective dimension values under the system, and fill the respective dimension value combinations and the calculated respective indicator values into the initial system data table to obtain a system data table; The risk definition module is used to confirm the dimension value combinations that need risk assessment under the selected system, set tax risks for the indicators that need risk assessment, and generate a risk rule table; wherein, the data in the risk rule table includes the abnormal judgment rules of the indicators and the dimension value combinations applicable to the rules; The risk list generation module is used to, when there is a corresponding dimension combination in the risk rule table, determine the abnormal indicator data in the system data table according to the abnormal judgment rules in the risk rule table, submit it to the review system, and generate a risk list after the review passes; A risk assessment report generation module, configured to generate a tax risk report based on a risk rule table, a risk list, and a system data table.

[0058] In some embodiments, the device further includes a dimension and index definition module, configured to define dimension information including a dimension number, a dimension name, and a dimension data source, and index information including an index number, an index name, and an index precision.

[0059] In some embodiments, the formula definition module includes a dimension value combination generation unit, a formula definition unit, an index data calculation unit, and a system data table generation unit; The dimension value combination generation unit is configured to select a system, calculate the Cartesian product of each dimension value in the system to generate all dimension value combinations; The formula definition unit is configured to, for each index in the system, determine the dimensions and calculation formulas on which the index calculation depends; The index data calculation unit is configured to traverse all dimension value combinations, substitute the dimension values in each combination into the calculation formulas of the corresponding indexes, and calculate the index values of each index under this combination; The system data table generation unit is configured to fill each dimension value combination and the calculated index values into the initial system data table to obtain a system data table.

[0060] In some embodiments, the formula definition unit includes a formula definition sub-module and a legality judgment sub-module; The formula definition sub-module is configured to use an index in the system as an output value, and all dimension information in the system or several other indexes in the same dimension system as input values; perform custom mathematical operations using the input values to construct a calculation formula for the output index; The legality judgment sub-module is configured to load the calculation formula of the output index, analyze all its dependent indexes, and the output index is the source index; judge the dependent indexes in turn; if there is a situation where a dependent index is an index data fetching and the dependent index is equal to the source index, it is determined as a circular dependency, and an illegal prompt message is returned; when all the dependent indexes of the source index have completed the parsing and circular judgment, if no circular dependency situation is found, the formula corresponding to the source index is legal.

[0061] Load the calculation formula of the output indicator, analyze all its dependent indicators including Indicator B and Indicator C, and the output indicator is the source indicator; judge the dependent indicators in turn; if Indicator B is a SQL data extraction, then continue to judge the next dependent indicator C; if Indicator C is an indicator data extraction, judge whether Indicator C is equal to the source indicator; if they are equal, it is determined as a circular dependency and an illegal prompt message is returned; if they are not equal, load the formula of Indicator C and analyze all its dependent indicators including Indicator D and Indicator A; if Indicator D is a SQL data extraction, skip it and judge whether Indicator A is equal to Source Indicator A; if they are equal, it is determined that there is a circular dependency and an illegal prompt message is returned; when all the dependent indicators of Indicator A have completed the parsing and circular judgment, if no circular dependency is found, the formula corresponding to Indicator A is legal.

[0062] In some embodiments, the risk definition module is specifically configured to confirm the dimension combination that needs to be risk-assessed under the selected system; analyze the indicator nature and business association of the indicators in the system, obtain the association between each indicator and tax risk, and determine the indicators that need to be risk-assessed; perform tax risk settings on the indicators that need to be risk-assessed and generate a risk rule table. Based on the analysis of the indicator nature and business association, determine the corresponding category for each indicator in the system; where the category includes positive indicators, inverse indicators, and moderate indicators; save the risk definition and synchronously generate a risk rule table.

[0063] In some embodiments, the device further includes a risk warning module, which is used to perform risk warning on the indicator data of the system data table according to the risk definition and the risk rule table, automatically submit the warning information for review, and transfer the risk warning information to the risk list after the review is passed.

[0064] In some embodiments, the risk assessment report generation module includes a collection and collation unit, a format conversion unit, a submission setting unit, and an acquisition unit; The collection and collation unit is used to integrate the exception judgment rules, dimension value combinations in the risk rule table, the exception indicator data in the risk list, and the dimension information and indicator information covered by the system data table; classify and summarize the exception indicator data in the risk list according to the exception judgment rules and risk types in the risk rule table; The format conversion unit is used to convert and adapt the integrated data and the classified and summarized data according to the format required by the large model; The submission setting unit is used to submit the adapted data to the large model through an interface or platform connected to the large model, set the relevant parameters of the large model according to the requirements of generating a tax risk report, and at the same time, input an instruction to generate a tax risk report based on the submitted data to the large model; An acquisition unit for acquiring a tax risk report generated by the large model after analyzing and processing the submitted data according to its own algorithms and training data after receiving the data and instructions.

[0065] An embodiment of the present invention further provides an electronic device, which includes: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The communication bus can be used for information transmission between the electronic device and the sensor. The processor can call the logical instructions in the memory to execute the following method: S1: Construct a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; S2: Select any system, calculate the indicator values under each combination according to the Cartesian product combination of the dimension values under this system, and fill the dimension value combinations and the indicator values into the initial system data table to obtain a system data table; S3: Confirm the dimension value combination that needs risk assessment, set tax risks for the indicators that need risk assessment, and generate a risk rule table; where the risk rule table contains an abnormal judgment rule and the dimension value combination applicable to this rule; S4: Determine the abnormal indicator data in the system data table according to the abnormal judgment rule in the risk rule table, submit it to the audit system, and generate a risk list after passing the audit; S5: Generate a tax risk report based on the risk rule table, the risk list, and the system data table.

[0066] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0067] An embodiment of the present invention provides a non-transitory computer-readable storage medium that stores computer instructions, and the computer instructions cause the computer to execute the method provided in the above method embodiment, for example, including: S1: constructing a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; S2: selecting any system, calculating the indicator values under each combination according to the Cartesian product combination of each dimension value under the system, and filling the dimension value combinations and the indicator values into the initial system data table to obtain a system data table; S3: confirming the dimension value combinations that need risk assessment, setting tax risks for the indicators that need risk assessment and generating a risk rule table; wherein, the risk rule table contains an abnormal judgment rule and the dimension value combinations applicable to the rule; S4: determining the abnormal indicator data in the system data table according to the abnormal judgment rule in the risk rule table, submitting it to the audit system, and generating a risk list after passing the audit; S5: generating a tax risk report based on the risk rule table, the risk list and the system data table.

[0068] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a tax risk assessment report, characterized in that: The steps include: S1: Build a system based on several dimensions and several indicators, and each system generates an initial system data table containing dimension information and indicator information; S2: Select any system, calculate the index values ​​under each combination according to the Cartesian product combination of each dimension value under the system, and fill the initial system data table with each dimension value combination and each index value to obtain a system data table; S3: Confirm the dimension value combination that needs risk assessment, set tax risks for the indicators that need risk assessment and generate a risk rule table; the risk rule table contains the abnormal judgment rule and the dimension value combination to which the rule applies; S4: Determine the abnormal indicator data in the system data table according to the abnormal judgment rules in the risk rule table, submit it to the review system, and generate a risk list after passing the review; S5: Generate a tax risk report based on the risk rule table, risk list and system data table.

2. The method for generating a tax risk assessment report according to claim 1, characterized in that: Step S1 includes: Dimension information and indicator information are defined, wherein the dimension information includes dimension number, dimension name and dimension data source, and the indicator information includes indicator number, indicator name and indicator precision.

3. The method for generating a tax risk assessment report according to claim 1, characterized in that: Step S2 includes: S21: Select any system, calculate the Cartesian product of each dimension value in the system to generate all dimension value combinations; S22: For each indicator in the system, determine the dimensions and calculation formulas on which the indicator calculation depends; S23: traverse all dimension value combinations, substitute the dimension value in each combination into the calculation formula of the corresponding indicator, and calculate the indicator value of each indicator under the combination; S24: Filling each dimension value combination and each indicator value obtained by calculation into the initial system data table to obtain a system data table.

4. The method for generating a tax risk assessment report according to claim 3, characterized in that: Step S22 includes: S221: taking an indicator in the system as an output value, and taking all dimensional information in the system or other indicators of the same dimensional system as input values; S222: constructing a calculation formula by performing a custom mathematical operation using the selected input values; S223: Load the calculation formula of the output indicator and analyze all its dependent indicators; determine the legality of the current formula, that is, determine whether there is a circular dependency of indicators; If the formula is legal, execute step S23; if the formula is illegal, execute step S222.

5. The method for generating a tax risk assessment report according to claim 4, characterized in that: Step S223 includes: S2231: Load the calculation formula of the output indicator and analyze all its dependent indicators. The output indicator is the source indicator; S2232: judging the dependent indicators in sequence; S2233: If there is a situation where the dependent indicator is an indicator fetch and the dependent indicator is equal to the source indicator, it is determined to be a circular dependency and an illegal prompt message is returned; S2234: When all dependent indicators of the source indicator have completed the analysis and circular judgment, if no circular dependency is found, the formula corresponding to the source indicator is legal.

6. The method for generating a tax risk assessment report according to claim 5, characterized in that: Step S3 includes: S31: Confirm the combination of dimensions that require risk assessment under the selected system; S32: Analyze the nature of the indicators in the system and their business relevance, obtain the relevance of each indicator to tax risk, and determine the indicators that require risk assessment; S33: Set tax risks for indicators that require risk assessment and generate a risk rule table.

7. The method for generating a tax risk assessment report according to claim 6, characterized in that: Step S5 includes: S51: Integrate the abnormal judgment rules and dimension value combinations in the risk rule table, the abnormal indicator data in the risk list, and the dimension information and indicator information covered by the system data table; S52: Classify and summarize the abnormal indicator data in the risk list according to the abnormal judgment rules and risk types in the risk rule table; S53: converting and adapting the integrated data and the classified and summarized data according to the format required by the large model; S54: Submit the adapted data to the big model through an interface or platform connected to the big model, set relevant parameters of the big model according to the requirements of generating a tax risk report, and input instructions for generating a tax risk report based on the submitted data into the big model; S55: After receiving the data and instructions, the big model generates a tax risk report by analyzing and processing the submitted data based on its own algorithm and training data.

8. A tax risk assessment report generating device, characterized in that: It includes system construction module, formula definition module, risk definition module, risk list generation module, and risk assessment report generation module; A system construction module is used to construct a system based on several dimensions and several indicators. Each system generates an initial system data table containing dimension information and indicator information. A formula definition module is used to select any system, calculate the index values ​​under each combination according to the Cartesian product combination of each dimension value under the system, and fill the initial system data table with each dimension value combination and each index value to obtain a system data table; The risk definition module is used to confirm the dimension value combinations that require risk assessment, set tax risks for the indicators that require risk assessment, and generate a risk rule table; the risk rule table contains the abnormal judgment rules and the dimension value combinations to which the rules apply; The risk list generation module is used to determine the abnormal indicator data in the system data table according to the abnormal judgment rules in the risk rule table when there is a corresponding dimension combination in the risk rule table, submit it to the review system, and generate a risk list after passing the review; The risk assessment report generation module is used to generate a tax risk report based on the risk rule table, risk list and system data table.

9. An electronic device, characterized in that: The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the tax risk assessment report generation method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the tax risk assessment report generating method as described in any one of claims 1 to 7.

Images