Tax refund declaration risk detection method and system based on big data, medium and equipment

Abstract

The invention relates to the technical field of business information processing, and provides a tax refund declaration risk detection method and system based on big data, a medium and equipment. Obtaining industrial data related to tax refund, screening out first data related to tax refund risk assessment from the industrial data according to the feature vector of the tax refund risk assessment target, and predicting the consumption of raw materials by using a yield ratio analysis model generated by pre-training, comparing the operation data, the sales data and the raw material consumption in the industrial data to generate a preliminary verification result of whether the tax refund declaration has a risk; according to the method, the tax refund declaration data are acquired, production areas and positioning data in the industrial data are compared with invoicing data in the tax refund declaration data, a logistics verification result is generated, abnormity and risk points in tax refund declaration are found in time, and finally a detailed risk assessment report is generated by integrating a preliminary verification result and the logistics verification result. Accurate identification and assessment of tax refund declaration risks are realized, and the accuracy and efficiency of risk assessment are improved.

Description

Technical Field

[0001] The present application relates to the field of commercial information processing technology, and more specifically, to a tax refund declaration risk detection method, system, medium and equipment based on big data. Background Art

[0002] At present, the risk detection of export tax rebate declaration mainly relies on manual review and simple numerical comparison. Tax regulatory authorities usually review the authenticity and rationality of tax rebate declarations based on the production and sales data reported by the enterprises themselves. However, this method has the problems of low audit efficiency and insufficient accuracy. With the complexity of international trade and the diversification of business models, traditional audit methods can no longer meet the needs of precise management.

[0003] At the same time, due to the lack of integration and analysis of multi-dimensional data from upstream and downstream of the industrial chain, existing technologies are unable to fully identify potential tax refund violations, which results in low risk detection efficiency and accuracy during the tax refund declaration process. Summary of the invention

[0004] The embodiments of the present application provide a tax refund declaration risk detection method, system, medium and device based on big data, which can at least to some extent solve the problem of low risk detection efficiency and accuracy in the tax refund declaration process.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to one aspect of the present application, a tax refund declaration risk detection method based on big data is provided, including: obtaining industrial data related to the tax refund, and screening out first data related to the tax refund risk assessment from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the positioning data of the transportation logistics, the business data, sales data and tax refund declaration data of the enterprise; the tax refund declaration data is input into a pre-trained and generated output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; the business data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration; the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result; and a risk assessment report for this tax refund is generated based on the preliminary verification result and the logistics verification result.

[0007] In the present application, based on the aforementioned scheme, the first data related to the tax refund risk assessment is screened out from the industry data according to the characteristic vector of the tax refund risk assessment target, including: standardizing the industry data to generate standard data in a unified form; generating a characteristic vector of the tax refund risk assessment target based on the industry data; performing feature selection from the standard data according to the characteristic vector of the tax refund risk assessment target to screen out the first data related to the tax refund risk assessment.

[0008] In the present application, based on the aforementioned scheme, before the tax refund declaration data is input into the pre-trained and generated cost-to-performance ratio analysis model to predict the raw material consumption corresponding to the industrial data, it also includes: generating a regression model based on the data structure of the tax refund declaration data, and determining the objective function corresponding to the regression model; tuning the objective function by cross-validation to determine the optimal regularization parameter of the objective function; iteratively updating the regression coefficient of the regression model by gradient descent, and using the updated regression model as the cost-to-performance ratio analysis model.

[0009] In the present application, based on the aforementioned scheme, the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration, including: classifying the raw material consumption according to the attributes of the raw materials, and determining the predicted consumption corresponding to each type of raw material; extracting the actual consumption corresponding to each type of raw material from the operating data and sales data in the industrial data; calculating the difference parameters between the predicted consumption and the actual consumption corresponding to each type of raw material, and generating a preliminary verification result of whether there is a risk in the tax refund declaration based on the difference parameters.

[0010] In the present application, based on the aforementioned scheme, the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result, including: extracting address-related target information from the invoicing data of the tax refund declaration data; comparing the production area and positioning data in the industrial data with the target data to generate a logistics verification result.

[0011] In the present application, based on the aforementioned scheme, a risk assessment report for this tax refund is generated according to the preliminary verification results and the logistics verification results, including: obtaining a preset report template; writing the preliminary verification results and the logistics verification results, as well as related data into the report template to generate a risk assessment report for this tax refund.

[0012] In the present application, based on the aforementioned solution, it also includes: querying the historical declaration information of the tax refund declaration subject corresponding to the industrial data; detecting whether there are risk items in the historical declaration information, and if so, writing the risk items into the risk assessment report.

[0013] According to one aspect of the present application, a tax refund declaration risk detection system based on big data is provided, comprising: An acquisition unit is used to acquire industrial data related to tax refund, and select first data related to tax refund risk assessment from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the location data of the transportation logistics, the business data of the enterprise, the sales data and the tax refund declaration data; A prediction unit, used for inputting the tax refund declaration data into a pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; A comparison unit, used to compare the operating data and sales data in the industrial data with the raw material consumption, and generate a preliminary verification result of whether there is a risk in the tax refund declaration; A verification unit, used to compare the production area and location data in the industrial data with the invoicing data in the tax refund declaration data to generate a logistics verification result; The reporting unit is used to generate a risk assessment report for this tax refund based on the preliminary verification result and the logistics verification result.

[0014] In the present application, based on the aforementioned scheme, the first data related to the tax refund risk assessment is screened out from the industry data according to the characteristic vector of the tax refund risk assessment target, including: standardizing the industry data to generate standard data in a unified form; generating a characteristic vector of the tax refund risk assessment target based on the industry data; performing feature selection from the standard data according to the characteristic vector of the tax refund risk assessment target to screen out the first data related to the tax refund risk assessment.

[0015] In the present application, based on the aforementioned scheme, before the tax refund declaration data is input into the pre-trained and generated cost-to-performance ratio analysis model to predict the raw material consumption corresponding to the industrial data, it also includes: generating a regression model based on the data structure of the tax refund declaration data, and determining the objective function corresponding to the regression model; tuning the objective function by cross-validation to determine the optimal regularization parameter of the objective function; iteratively updating the regression coefficient of the regression model by gradient descent, and using the updated regression model as the cost-to-performance ratio analysis model.

[0016] In the present application, based on the aforementioned scheme, the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration, including: classifying the raw material consumption according to the attributes of the raw materials, and determining the predicted consumption corresponding to each type of raw material; extracting the actual consumption corresponding to each type of raw material from the operating data and sales data in the industrial data; calculating the difference parameters between the predicted consumption and the actual consumption corresponding to each type of raw material, and generating a preliminary verification result of whether there is a risk in the tax refund declaration based on the difference parameters.

[0017] In the present application, based on the aforementioned scheme, the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result, including: extracting address-related target information from the invoicing data of the tax refund declaration data; comparing the production area and positioning data in the industrial data with the target data to generate a logistics verification result.

[0018] In the present application, based on the aforementioned scheme, a risk assessment report for this tax refund is generated according to the preliminary verification results and the logistics verification results, including: obtaining a preset report template; writing the preliminary verification results and the logistics verification results, as well as related data into the report template to generate a risk assessment report for this tax refund.

[0019] In the present application, based on the aforementioned solution, it also includes: querying the historical declaration information of the tax refund declaration subject corresponding to the industrial data; detecting whether there are risk items in the historical declaration information, and if so, writing the risk items into the risk assessment report.

[0020] According to one aspect of the present application, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the tax refund declaration risk detection method based on big data as described in the above embodiment is implemented.

[0021] According to one aspect of the present application, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the tax refund declaration risk detection method based on big data as described in the above embodiments.

[0022] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the tax refund declaration risk detection method based on big data provided in the above-mentioned various optional implementations.

[0023] In the technical solution of this application, the accuracy and pertinence of the data are ensured by accurately screening the industrial data related to the tax refund risk assessment, and the raw material consumption is predicted by using the pre-trained output-to-cost analysis model. By comparing the operation and sales data with the raw material consumption, as well as the regional and positioning data with the invoicing data, the anomalies and risk points in the tax refund declaration can be discovered in time. Finally, the preliminary verification and logistics verification results are combined to generate a detailed risk assessment report, which realizes the accurate identification and assessment of the tax refund declaration risks and improves the accuracy and efficiency of risk assessment.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 The flowchart of a tax refund declaration risk detection method based on big data in one embodiment of the present application is schematically shown.

[0027] Figure 2 The flowchart of screening data in one embodiment of the present application is schematically shown.

[0028] Figure 3 A schematic diagram of a tax refund declaration risk detection system based on big data in one embodiment of the present application is schematically shown.

[0029] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0030] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0031] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0032] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0033] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0034] The implementation details of the technical solution of this application are described in detail below: Figure 1 The flowchart of the tax refund declaration risk detection method based on big data according to an embodiment of the present application is shown. Figure 1 As shown, the tax refund declaration risk detection method based on big data includes at least steps S110 to S150, which are described in detail as follows: In step S110, industrial data related to tax refund is obtained, and first data related to tax refund risk assessment is screened out from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the positioning data of the transportation logistics, the business data, sales data and tax refund declaration data of the enterprise.

[0035] In the embodiment of the present application, in order to conduct tax refund risk assessment, it is first necessary to obtain industry data related to tax refund, which covers the import and export records, financial status, tax records, etc. of the enterprise. Then, according to the characteristic vector of the tax refund risk assessment target, the first data with high correlation with the tax refund risk assessment is accurately identified and extracted from the huge industry data, laying a solid foundation for the subsequent risk assessment work.

[0036] Specifically, in this embodiment, the production area of ​​the product includes the final production site, assembly site, and processing site of the product; the positioning data of transportation logistics includes the geographical location and transportation route of the goods, etc. Through the global navigation system and sensor technology, the logistics system can obtain these positioning data in real time; the business data of the enterprise includes various data generated by the enterprise in daily business activities, such as sales, costs, profits, human resources, and market share; sales data includes various data generated by the enterprise in sales activities, including sales date, sales area, sales channel, product series, product name, product price, sales and sales volume; tax refund declaration data includes relevant data and information that the enterprise needs to submit when applying for tax refund, including purchase details declaration data and export details declaration data. Among them, the purchase details declaration data mainly records the information of raw materials, parts and components purchased by the enterprise, including the name, quantity, unit price and supplier of the goods. The export details declaration data mainly records the information of the enterprise's exported goods, including the name, quantity, export date and export country of the goods.

[0037] like Figure 2 As shown, in one embodiment of the present application, first data related to the tax refund risk assessment is screened out from the industry data according to the feature vector of the tax refund risk assessment target, including: S210, performing standardization processing on the industrial data to generate standard data in a unified format; S220, generating a feature vector of a tax refund risk assessment target based on the industry data; S230, performing feature selection from the standard data according to the feature vector of the tax refund risk assessment target, and screening out first data associated with the tax refund risk assessment.

[0038] After obtaining the industry data related to tax refund, the industry data is standardized to generate standard data corresponding to the industry data. Dat_snd_i , standard data Dat_snd_i for: in, Dat_org_i is the original data, i is the data identifier, It is the transformation parameter generated by the maximum likelihood estimation of industrial data. The industrial data is processed by the above standardization method to generate standard data in a unified form.

[0039] After generating standard data, based on industry data, such as the company's production cost, total production, raw material procurement data, sales data, etc., a feature vector corresponding to the tax refund risk assessment target is generated to measure the company's tax refund risk.

[0040] Afterwards, feature selection is performed from the standard data according to the feature vector corresponding to the tax refund risk assessment target, and the correlation parameter between each standard data and the tax refund risk assessment is calculated. Cnn_i , correlation parameter Cnn_i for: in, and j Indicates data identification, n Indicates the amount of data, Indicates i The characteristic vector of the tax refund risk assessment target corresponding to the data, They are i and j The characteristic coefficient vector corresponding to each data, which is used to represent the weight of each feature in the tax refund risk assessment; It is i The feature vector of the data, is the regularization parameter, p is the number of features.

[0041] Determine the feature vector of the current data through the above process The characteristic vector of the tax refund risk assessment target The correlation parameter between them is then determined according to the value of the correlation parameter, and the feature vector closely related to the tax refund risk assessment target is determined, so that the first data related to the tax refund risk assessment can be determined. For example, the data corresponding to the feature whose correlation parameter value is greater than or equal to the set threshold is selected as the first data related to the tax refund risk assessment.

[0042] The above process obtains industrial data related to tax refunds, standardizes the industrial data, generates standard data in a unified form, and screens out relevant industrial data from the standard data based on the characteristic vector of the tax refund risk assessment target. It can efficiently integrate and utilize a large amount of data resources, ensure the data relevance and accuracy of subsequent analysis, reduce the interference of noise data, and improve the accuracy of risk assessment.

[0043] In step S120, the tax refund declaration data is input into a pre-trained and generated production-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data.

[0044] In this embodiment, the tax refund declaration data, including detailed purchase and export details, is input into a production-to-cost ratio analysis model that is pre-trained with a large amount of historical data. The model can use advanced algorithms and machine learning techniques to accurately predict the raw material consumption required in the actual production process of the industry corresponding to these tax refund declaration data.

[0045] In one embodiment of the present application, before inputting the tax refund declaration data into a pre-trained and generated output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data, the method further includes: Generate a regression model based on the data structure of the tax refund declaration data, and determine the objective function corresponding to the regression model; The objective function is tuned by cross-validation to determine the optimal regularization parameter of the objective function; The regression coefficient of the regression model is iteratively updated by means of gradient descent, and the regression model obtained after the update is used as the cost-effectiveness analysis model.

[0046] In one embodiment of the present application, previously processed tax refund and production related data are obtained as historical data, and characteristic variables and target variables are extracted from the historical data. The characteristic variables may include product type, quantity, price, region, mode of transportation, etc., and the target variable is the consumption of raw materials. A regression model is generated based on the data structure of the tax refund declaration data, and the target function corresponding to the regression model is determined. Obj , the objective function Obj for: in, n represents the sample size; and j Indicates data identification, Indicates i The target variable value of samples; Represents the regression coefficient, which is used to measure the influence of the feature variable on the target variable. represents the initial value of the regression coefficient; p represents the number of features; Representative i The sample j eigenvalues; represents the regularization parameter, It is a penalty term coefficient introduced into the objective function.

[0047] In this embodiment, the optimal Specifically, the dataset is divided into a training set and a validation set to evaluate the performance of the model on different subsets. For candidate values, the model is trained on the training set and its performance is evaluated on the validation set. Too large a value may result in a model that is too simple and cannot capture the complex structure of the data; If the value is too small, the model may be too complex and prone to overfitting on the training data. The value of can control the complexity of the model and prevent the model from overfitting or underfitting. Finally, the model with the best performance on the validation set is selected. The value is taken as the final optimal regularization parameter.

[0048] In this embodiment, the regression coefficient are parameters in the model that determine the degree of influence of the feature variable on the target variable. Before model training begins, set an initial value for these parameters . Regression coefficient The initial value of can be set to zero or a random decimal. Zero initialization means that all feature variables have the same effect on the target variable at the beginning of training; while random initialization gives each feature variable a random starting effect. The choice of initialization method may affect the training efficiency and final performance of the model. In some cases, random initialization helps the model find the optimal solution faster.

[0049] After that, the regression coefficient of the regression model is iteratively updated by gradient descent, and the parameter value is updated according to the gradient of the objective function. In each iteration, the gradient of the objective function to the current parameter is calculated, and the parameter value is updated in the opposite direction of the gradient. Through multiple iterations, the minimum value of the objective function can be gradually approached, and the regression model obtained after the update is used as the cost-effectiveness analysis model.

[0050] Before the application of the cost-to-performance ratio analysis model, the accuracy and robustness of the model were ensured by generating a regression model, tuning the objective function and iteratively updating the regression coefficients, so that the cost-to-performance ratio analysis model can more accurately predict the raw material consumption and provide strong support for tax refund risk assessment.

[0051] In this embodiment, the raw material consumption is predicted using a pre-trained output-to-cost analysis model. This step can perform in-depth mining and analysis of the tax refund declaration data based on historical data and business logic.

[0052] In step S130, the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration.

[0053] The operating data and sales data in the collected industrial data are compared and analyzed with the raw material consumption predicted by the output-to-cost ratio analysis model. The purpose is to check whether the business sales of the enterprise are reasonably consistent with the predicted raw material consumption. Through this comparison, the system can automatically generate preliminary verification results on whether there are potential risks in the tax refund declaration, and can preliminarily determine whether there are abnormalities in the tax refund declaration, such as false production, exaggerated costs and other risks.

[0054] In one embodiment of the present application, the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration, including: Classify the raw material consumption according to the properties of the raw materials and determine the corresponding predicted consumption of each type of raw material; Extracting the actual consumption corresponding to each type of raw material from the operating data and sales data in the industrial data; The difference parameters between the predicted consumption and the actual consumption corresponding to each type of raw material are calculated, and a preliminary verification result of whether there is a risk in the tax refund declaration is generated based on the difference parameters.

[0055] In one embodiment of the present application, deep learning is used to automatically classify raw materials. By analyzing the properties of raw materials (such as material, purpose, cost, etc.) and historical data, the corresponding types of raw materials are automatically learned and identified, and then raw materials belonging to the same type are determined. The raw material consumption is classified, and the predicted consumption corresponding to each type of raw material is determined from the raw material consumption.

[0056] Through natural language processing and regular expression technologies, information related to raw material consumption is automatically extracted from business data and sales data as actual consumption. The extracted data is aggregated to form a data set of actual consumption of various raw materials. Afterwards, data accuracy and completeness can be ensured through data verification.

[0057] Calculate the difference parameters of various raw materials based on the predicted consumption and actual consumption data sets. Perform real-time analysis on the difference parameters. If the difference parameter is greater than or equal to the preset threshold, it is determined as a potential risk point, and a preliminary verification result is generated to determine whether there is a risk in the tax refund declaration.

[0058] Optionally, when the difference parameter exceeds a preset threshold, the computer automatically triggers an early warning mechanism and generates a risk early warning report.

[0059] In step S140, the production area and location data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result.

[0060] The production area and location data in the industrial data are carefully compared and analyzed with the invoicing data recorded in the tax refund declaration data. This process focuses on verifying the consistency between the actual production location and transportation route of the product and the invoicing information, aiming to detect whether there are any anomalies or inconsistencies in logistics. Through this comparison, the system can automatically generate logistics verification results, provide logistics compliance verification for the tax refund review process, and ensure the accuracy and rigor of the tax refund operation.

[0061] In one embodiment of the present application, the production area and location data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result, including: Extracting target information related to addresses from invoicing data in tax refund declaration data; The production area and location data in the industrial data are compared with the target data to generate a logistics verification result.

[0062] In one embodiment of the present application, an address resolution service based on natural language processing is applied to resolve the address information in the invoicing data, identify and extract key information in the address, such as country / region, province, city, street, etc.

[0063] The address information in the extracted invoicing data is compared with the production area in the industrial data, for example, by string matching, fuzzy matching or geographic location-based matching methods, similarity data is calculated, and based on the comparison result between the similarity data and the preset threshold, it is determined whether the address information in the invoicing data matches the production area in the industrial data. Based on the comparison result, a logistics verification result is generated. If the comparison result is that the similarity data is greater than or equal to the set threshold, the generated logistics verification result is passed.

[0064] Specifically, the logistics verification results may include a list of successfully matched records, analysis of the reasons for mismatches (such as inconsistent address information, missing industry data, etc.), and possible logistics risks (such as high cross-border logistics costs, long transportation time, etc.).

[0065] Furthermore, for successfully matched records, more detailed logistics analysis can be performed in combination with positioning data, such as calculating transportation distance, evaluating transportation costs, etc.

[0066] The above process can comprehensively detect potential risks in tax refund declarations by classifying and comparing raw material consumption, as well as comparing production area and location data with invoicing data. Preliminary verification can identify anomalies in cost or output, while logistics verification can verify the authenticity and rationality of transactions, thereby effectively preventing false declarations.

[0067] In step S150, a risk assessment report for this tax refund is generated based on the preliminary verification result and the logistics verification result.

[0068] Based on the analysis of the comparison between the enterprise's operating sales data and raw material consumption in the preliminary verification results, and the verification of the consistency of the production area, positioning data and invoicing data in the logistics verification results, and comprehensive consideration of the matching degree of various data and potential inconsistencies, a risk assessment report for this tax refund application is finally generated based on the preliminary verification results and logistics verification results.

[0069] In one embodiment of the present application, a risk assessment report for this tax refund is generated based on the preliminary verification result and the logistics verification result, including: Get preset report templates; The preliminary verification results, the logistics verification results, and related data are written into the report template to generate a risk assessment report for this tax refund.

[0070] In one embodiment of the present application, a dedicated template storage area is set up to store and manage various preset report templates. According to user instructions or preset business rules, the computer selects a suitable report template from the template storage area, loads the template into the memory, and prepares for subsequent data filling operations.

[0071] Optionally, the user or system administrator is allowed to customize the template, such as setting the report title, header and footer, font style, etc.

[0072] Organize preliminary verification results, logistics verification results and related data into a structured data format to ensure the accuracy and completeness of the data, and avoid data errors or omissions through data verification and exception handling mechanisms.

[0073] Parse the loaded report template and identify the placeholders or data fields in the report template. Establish a mapping relationship between the data fields and the template placeholders to ensure that the data can be correctly filled into the corresponding positions in the report template. According to the mapping relationship, fill the prepared data into the placeholder positions in the report template. Format the data, such as date formatting, number formatting, etc., to ensure that the data is presented in the correct format in the template.

[0074] Introduce data visualization technology, such as charts, graphs, and dashboards, to display complex data in an intuitive way in the report. After completing data filling and formatting, generate a risk assessment report for this tax refund. By optimizing the report generation process and improving data visualization capabilities, the quality and value of the report can be significantly improved.

[0075] The above process generates a risk assessment report by writing the preliminary verification results, logistics verification results and related data into the preset report template. It not only provides intuitive assessment results, but also facilitates the review and management of tax refund declarations by relevant personnel. The preset report template also ensures the standardization and consistency of the report and improves work efficiency.

[0076] In one embodiment of the present application, it also includes: Query the historical declaration information of the tax refund declaration entity corresponding to the industrial data; Check whether there are risk items in the historical declaration information, and if so, write the risk items into the risk assessment report.

[0077] In one embodiment of the present application, a centralized database or data warehouse is pre-established to store the historical declaration information of all tax refund declaration entities. A user-friendly query interface or API interface is designed to allow users to enter query conditions and support paging query and sorting functions to improve query efficiency and user experience.

[0078] Record historically detected risk items in the database for subsequent analysis and processing. Ensure the uniqueness and traceability of risk item records, and identify and record them through timestamps, operation logs, etc. Extract risk item records related to the current tax refund declaration entity from the database. Based on the risk item records, fill in the template of the risk assessment report and generate a report containing risk items and their analysis.

[0079] This embodiment can achieve comprehensive monitoring and risk assessment of the tax refund declaring entity by querying the historical declaration information of the tax refund declaring entity corresponding to the industrial data and detecting whether there are risk items therein. Writing historical risk items into the risk assessment report helps relevant personnel to have a more comprehensive understanding of the credit status and risk level of the declaring entity, providing an important reference for future tax refund approval.

[0080] In the technical solution of the present application, the industrial data related to the tax refund is obtained, and the first data related to the tax refund risk assessment is screened out from the industrial data according to the characteristic vector of the tax refund risk assessment target; the tax refund declaration data is input into the pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration; the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result; based on the preliminary verification result and the logistics verification result, a risk assessment report for this tax refund is generated. By accurately screening the industrial data related to the tax refund risk assessment, the accuracy and pertinence of the data are ensured, and the pre-trained output-to-cost ratio analysis model is used to predict the raw material consumption. By comparing the operating and sales data with the raw material consumption, as well as the regional and positioning data with the invoicing data, the anomalies and risk points in the tax refund declaration can be discovered in a timely manner. Finally, the preliminary verification and logistics verification results are integrated to generate a detailed risk assessment report, which enables accurate identification and assessment of tax refund declaration risks and improves the accuracy and efficiency of risk assessment.

[0081] The following describes an embodiment of the device of the present application, which can be used to execute the tax refund declaration risk detection method based on big data in the above embodiment of the present application. It can be understood that the device can be a computer program (including program code) running in a computer device, for example, the device is an application software; the device can be used to execute the corresponding steps in the method provided in the embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the tax refund declaration risk detection method based on big data in the above embodiment of the present application.

[0082] Figure 3 A block diagram of a tax refund declaration risk detection system based on big data according to an embodiment of the present application is shown.

[0083] Reference Figure 3 As shown, a tax refund declaration risk detection system based on big data according to an embodiment of the present application includes: The acquisition unit 310 is used to acquire industrial data related to tax refund, and select first data related to tax refund risk assessment from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the location data of the transportation logistics, the business data, sales data and tax refund declaration data of the enterprise; The prediction unit 320 is used to input the tax refund declaration data into a pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; A comparison unit 330 is used to compare the operating data and sales data in the industrial data with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration; A verification unit 340 is used to compare the production area and location data in the industrial data with the invoicing data in the tax refund declaration data to generate a logistics verification result; The reporting unit 350 is used to generate a risk assessment report for this tax refund based on the preliminary verification result and the logistics verification result.

[0084] In the present application, based on the aforementioned scheme, the first data related to the tax refund risk assessment is screened out from the industry data according to the characteristic vector of the tax refund risk assessment target, including: standardizing the industry data to generate standard data in a unified form; generating a characteristic vector of the tax refund risk assessment target based on the industry data; performing feature selection from the standard data according to the characteristic vector of the tax refund risk assessment target to screen out the first data related to the tax refund risk assessment.

[0085] In the present application, based on the aforementioned scheme, before the tax refund declaration data is input into the pre-trained and generated cost-to-performance ratio analysis model to predict the raw material consumption corresponding to the industrial data, it also includes: generating a regression model based on the data structure of the tax refund declaration data, and determining the objective function corresponding to the regression model; tuning the objective function by cross-validation to determine the optimal regularization parameter of the objective function; iteratively updating the regression coefficient of the regression model by gradient descent, and using the updated regression model as the cost-to-performance ratio analysis model.

[0086] In the present application, based on the aforementioned scheme, the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration, including: classifying the raw material consumption according to the attributes of the raw materials, and determining the predicted consumption corresponding to each type of raw material; extracting the actual consumption corresponding to each type of raw material from the operating data and sales data in the industrial data; calculating the difference parameters between the predicted consumption and the actual consumption corresponding to each type of raw material, and generating a preliminary verification result of whether there is a risk in the tax refund declaration based on the difference parameters.

[0087] In the present application, based on the aforementioned scheme, the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result, including: extracting address-related target information from the invoicing data of the tax refund declaration data; comparing the production area and positioning data in the industrial data with the target data to generate a logistics verification result.

[0088] In the present application, based on the aforementioned scheme, a risk assessment report for this tax refund is generated according to the preliminary verification results and the logistics verification results, including: obtaining a preset report template; writing the preliminary verification results and the logistics verification results, as well as related data into the report template to generate a risk assessment report for this tax refund.

[0089] In the present application, based on the aforementioned solution, it also includes: querying the historical declaration information of the tax refund declaration subject corresponding to the industrial data; detecting whether there are risk items in the historical declaration information, and if so, writing the risk items into the risk assessment report.

[0090] In the technical solution of the present application, the industrial data related to the tax refund is obtained, and the first data related to the tax refund risk assessment is screened out from the industrial data according to the characteristic vector of the tax refund risk assessment target; the tax refund declaration data is input into the pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; the operating data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration; the production area and positioning data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result; based on the preliminary verification result and the logistics verification result, a risk assessment report for this tax refund is generated. By accurately screening the industrial data related to the tax refund risk assessment, the accuracy and pertinence of the data are ensured, and the pre-trained output-to-cost ratio analysis model is used to predict the raw material consumption. By comparing the operating and sales data with the raw material consumption, as well as the regional and positioning data with the invoicing data, the anomalies and risk points in the tax refund declaration can be discovered in a timely manner. Finally, the preliminary verification and logistics verification results are integrated to generate a detailed risk assessment report, which enables accurate identification and assessment of tax refund declaration risks and improves the accuracy and efficiency of risk assessment.

[0091] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown.

[0092] It should be noted that the computer system of the electronic device in this embodiment is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0093] In this embodiment, the computer system includes a central processing unit 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory 402 or the program loaded from the storage part 408 to the random access memory 403, such as executing the tax refund declaration risk detection method based on big data described in the above embodiment. In the random access memory 403, various programs and data required for system operation are also stored. The central processing unit 401, the read-only memory 402 and the random access memory 403 are connected to each other through the bus 404. The input / output interface 405 is also connected to the bus 404.

[0094] The following components are connected to the input / output interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0095] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 409, and / or installed from a removable medium 411. When the computer program is executed by the central processing unit 401, various functions defined in the system of the present application are executed.

[0096] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0097] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0098] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0099] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the above-mentioned various optional implementations.

[0100] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the tax refund declaration risk detection method based on big data described in the above embodiment.

[0101] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0102] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the implementation method of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation method of the present application.

[0103] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0104] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A tax refund declaration risk detection method based on big data, characterized in that: include: Acquire industrial data related to tax refund, and select first data related to tax refund risk assessment from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the location data of the transportation logistics, the business data, the sales data and the tax refund declaration data of the enterprise; Input the tax refund declaration data into a pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; Compare the operating data and sales data in the industrial data with the raw material consumption to generate a preliminary verification result on whether there is a risk in the tax refund declaration; Compare the production area and location data in the industrial data with the invoicing data in the tax refund declaration data to generate a logistics verification result; A risk assessment report for this tax refund is generated based on the preliminary verification results and the logistics verification results.

2. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: The step of selecting first data related to the tax refund risk assessment from the industry data according to the characteristic vector of the tax refund risk assessment target includes: Standardizing the industrial data to generate standard data in a unified form; Based on the industrial data, generating a characteristic vector of a tax refund risk assessment target; Feature selection is performed from the standard data according to the feature vector of the tax refund risk assessment target to screen out first data associated with the tax refund risk assessment.

3. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: Before inputting the tax refund declaration data into a pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data, the method further includes: Generate a regression model based on the data structure of the tax refund declaration data, and determine the objective function corresponding to the regression model; The objective function is tuned by cross-validation to determine the optimal regularization parameter of the objective function; The regression coefficient of the regression model is iteratively updated by means of gradient descent, and the regression model obtained after the update is used as the cost-effectiveness analysis model.

4. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: The operation data and sales data in the industrial data are compared with the raw material consumption to generate a preliminary verification result of whether there is a risk in the tax refund declaration, including: Classifying the raw material consumption according to the properties of the raw materials corresponding to the raw material consumption, and determining the predicted consumption corresponding to each type of raw material; Extracting the actual consumption corresponding to each type of raw material from the operating data and sales data in the industrial data; The difference parameters between the predicted consumption and the actual consumption corresponding to each type of raw material are calculated, and a preliminary verification result of whether there is a risk in the tax refund declaration is generated based on the difference parameters.

5. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: The production area and location data in the industrial data are compared with the invoicing data in the tax refund declaration data to generate a logistics verification result, including: Extract the target data related to the address from the invoicing data of the tax refund declaration data; The production area and location data in the industrial data are compared with the target data to generate a logistics verification result.

6. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: The risk assessment report for this tax refund is generated based on the preliminary verification results and the logistics verification results, including: Get preset report templates; The preliminary verification results and the logistics verification results are written into the report template to generate a risk assessment report for this tax refund.

7. The tax refund declaration risk detection method based on big data according to claim 1 is characterized in that: The method further comprises: Query the historical declaration information of the tax refund declaration entity corresponding to the industrial data; Check whether there are risk items in the historical declaration information, and if so, write the risk items into the risk assessment report.

8. A tax refund declaration risk detection system based on big data, characterized in that: include: An acquisition unit is used to acquire industrial data related to tax refund, and select first data related to tax refund risk assessment from the industrial data according to the characteristic vector of the tax refund risk assessment target; the industrial data includes the production area of ​​the product, the location data of the transportation logistics, the business data of the enterprise, the sales data and the tax refund declaration data; A prediction unit, used for inputting the tax refund declaration data into a pre-trained output-to-cost ratio analysis model to predict the raw material consumption corresponding to the industrial data; A comparison unit, used to compare the operating data and sales data in the industrial data with the raw material consumption, and generate a preliminary verification result of whether there is a risk in the tax refund declaration; A verification unit, used to compare the production area and location data in the industrial data with the invoicing data in the tax refund declaration data to generate a logistics verification result; The reporting unit is used to generate a risk assessment report for this tax refund based on the preliminary verification result and the logistics verification result.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the tax refund declaration risk detection method based on big data as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the tax refund declaration risk detection method based on big data as described in any one of claims 1 to 7.

Images