Cross-mechanism flexible data report generation method and system

Through the combination of deep learning models and historical report data, the accuracy and completeness problems in report generation are solved, effective verification and quality assurance of report data are achieved, and data quality of the direct reporting platform is ensured.

CN119047452AInactive Publication Date: 2024-11-29TONGXIANG WUJIANG TECH DEV CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411524864.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-30
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the prior art, reports generated by designated templates are prone to accuracy and completeness problems during data entry, and the rationality of the report cannot be effectively verified, affecting the quality of the report uploaded to the direct reporting platform.

Method used

The cross-organization flexible data report generation method is adopted, and the report data is processed using deep learning model. The numerical deviation value of key data types is calculated based on historical report data, and the target report is generated when the deviation value is lower than the threshold to ensure the accuracy of the report and Integrity.

Benefits of technology

It improves the accuracy and completeness of reports, ensures the quality of data uploaded to the direct reporting platform, and provides a cross-institutional data verification method to ensure the rationality and reliability of the data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047452A_ABST
    Figure CN119047452A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of report data processing, and discloses a cross-mechanism flexible data report generation method and system, and the method comprises the steps: authenticating a first node based on a report generation request, inputting report data to a deep learning model after the authentication is passed, and extracting a report data type and a corresponding numerical value; all the numerical values can be imported into the data template, so that when it is detected that no numerical value of the key data type exists, the numerical value of the key data type is effectively and rapidly calculated in combination with all the numerical values; in addition, a second node can be determined in combination with the initial organization name of the first node, so that the numerical deviation value of the key data type is obtained according to the first historical report of the first node and the second historical report of the second node, and the reasonability of the generated data is verified in a cross-organization data verification mode. And the target report is generated when the numerical value deviation value is lower than the preset threshold value, so that the accuracy and integrity of the report are improved, and the quality of the report uploaded to the direct report platform can be guaranteed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of report data processing, and in particular, relates to a method and system for generating cross-institutional flexible data reports. Background Art

[0002] When enterprises or institutions are aggregating data, in order to avoid tedious statistical reporting, they can generally extract data such as inventory data, sales data, or invoice data from the internal system, and import the data into the specified template to generate reports. The generated reports can also be submitted to the direct reporting platform according to statistical survey needs.

[0003] Since the data in the internal system is mainly entered manually, when there are problems such as missing data or data anomalies due to manual operations, it is easy to cause the accuracy and completeness of the reports generated by the specified template to be poor; and although conventional data processing methods can fill in or correct data to a certain extent, they cannot effectively verify the rationality of the generated reports, thereby affecting the quality of reports uploaded to the direct reporting platform. Summary of the invention

[0004] In order to solve the technical problems that the accuracy and completeness of reports generated by designated templates are poor, and the conventional data processing method cannot effectively verify the rationality of the generated reports, thereby affecting the quality of reports uploaded to the direct reporting platform, this application proposes a cross-institutional flexible data report generation method and system, and its technical solution is as follows: In a first aspect, an embodiment of the present application provides a method for generating a cross-institutional flexible data report, comprising: When receiving a report generation request from the first node, performing authentication processing on the first node based on the report generation request, and after determining that the first node passes the authentication, obtaining the report data uploaded by the first node; When the report data is identified as an image file, the report data is input into a deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and report data labels marked by each report sample image, and each report data label includes at least two standard data types and a numerical value corresponding to each standard data type; Import the values ​​corresponding to all report data types into the data template, and when it is detected that the value corresponding to the key data type does not exist in the data template after the import processing, calculate the value corresponding to the key data type based on the values ​​corresponding to all report data types; Determine the second node according to the initial organization name configured by the first node, and obtain the first historical report corresponding to the first node and the second historical report corresponding to the second node; Based on the first historical report and the second historical report, a numerical deviation value corresponding to the key data type is determined, and when the numerical deviation value is lower than a preset threshold, a target report is generated based on the imported processed data template, the numerical value corresponding to the key data type and the first historical report.

[0005] In an optional solution of the first aspect, performing authentication processing on the first node based on the report generation request includes: Parsing the requesting organization name from the report generation request, and when detecting that the initial organization name configured by the first node is inconsistent with the requesting organization name, issuing a verification request according to the user information configured by the first node; When confirmation information of the verification request is received, determining that the first node is authenticated; or When it is detected that the initial organization name configured by the first node is consistent with the requesting organization name, it is determined that the first node is authenticated.

[0006] In yet another optional solution of the first aspect, before inputting the report data into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, after identifying that the report data is an image file, the method further includes: Performing image enhancement processing on the report data, and marking characteristic pattern contours in the report data after image enhancement processing; Identify at least two edge pixel values ​​of the characteristic pattern contour, and determine whether the characteristic pattern contour has edge anomalies based on all edge pixel values; The report data is input into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, including: When it is determined that there is no edge anomaly in the characteristic pattern contour, the report data is input into the deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type.

[0007] In another optional solution of the first aspect, after acquiring the report data uploaded by the first node, the method further includes: When the report data is identified as a text file, the report data is standardized; Based on a preset data type library, the standardized report data is matched to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the preset data type library includes at least two standard data types and a synonymous data type corresponding to each standard data type.

[0008] In yet another optional solution of the first aspect, the key data type is product output value; Calculate the values ​​corresponding to the key data types based on the values ​​corresponding to all report data types, including: When all report data types include product output, invoiced quantity and invoiced amount, the numerical value corresponding to the product output, the numerical value corresponding to the invoiced quantity and the numerical value corresponding to the invoiced amount are substituted into the preset first formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output and sales unit price, the numerical value corresponding to the product output and the numerical value corresponding to the sales unit price are substituted into the preset second formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output, determine the value corresponding to the invoiced quantity of at least one sub-product and the value corresponding to the invoiced amount of each sub-product according to the product name corresponding to the product output; The values ​​corresponding to the invoiced quantities of all sub-products are summed to obtain a first summation result, and the values ​​corresponding to the invoiced amounts of all sub-products are summed to obtain a second summation result; The first summation result, the second summation result and the product output are substituted into a preset third formula to calculate a numerical value corresponding to the product output value.

[0009] In yet another optional solution of the first aspect, determining the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report includes: At a preset time interval, at least two first values ​​corresponding to the product output value are selected from the first historical report, and a fitting process is performed on all the first values ​​and the values ​​corresponding to the product output value to obtain a first fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent first moments and the numerical value corresponding to each first moment are determined in the first fitting curve, and the first numerical value change value is calculated according to all the first moments and the numerical value corresponding to each first moment; At a preset time interval, at least two second values ​​corresponding to the product output value are selected from the second historical report, and a fitting process is performed on all the second values ​​to obtain a second fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent second moments and the numerical value corresponding to each second moment are determined in the second fitting curve, and the second numerical value change value is calculated according to all the second moments and the numerical value corresponding to each second moment; The difference between the first numerical change value and the second numerical change value is calculated to obtain a numerical deviation value corresponding to the key data type.

[0010] In another optional solution of the first aspect, generating a target report according to the imported processed data template, the numerical value corresponding to the key data type, and the first historical report includes: Summing all the first values ​​and the values ​​corresponding to the product output values ​​to obtain a third summation result; Generate a target report based on the imported processed data template, the numerical value corresponding to the product output value and the third summation result.

[0011] In a second aspect, an embodiment of the present application provides a cross-institutional flexible data report generation system, including: A node authentication module, configured to, when receiving a report generation request from a first node, perform authentication processing on the first node based on the report generation request, and after determining that the first node has passed the authentication, obtain the report data uploaded by the first node; A data extraction module, for inputting the report data into a deep learning model when identifying the report data as an image file, to obtain at least two report data types and numerical values ​​corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and report data labels marked by each report sample image, and each report data label includes at least two standard data types and numerical values ​​corresponding to each standard data type; A numerical calculation module is used to import the numerical values ​​corresponding to all report data types into the data template, and when it is detected that the numerical value corresponding to the key data type does not exist in the data template after the import processing, calculate the numerical value corresponding to the key data type based on the numerical values ​​corresponding to all report data types; A data query module, used to determine the second node according to the initial organization name configured by the first node, and obtain a first historical report corresponding to the first node and a second historical report corresponding to the second node; A report generation module is used to determine the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report, and when the numerical deviation value is lower than a preset threshold, generate a target report based on the imported processed data template, the numerical value corresponding to the key data type and the first historical report.

[0012] In a third aspect, an embodiment of the present application further provides a cross-organization flexible data report generation system, including a processor and a memory; The processor is connected to the memory; A memory for storing executable program codes; The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the cross-institutional flexible data report generation method provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the cross-institutional flexible data report generation method provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect can be implemented.

[0014] In an embodiment of the present application, the first node can be enterprise authenticated based on the report generation request, and the uploaded report data can be input into the deep learning model after the authentication is passed, so as to more accurately extract various report data types and corresponding values ​​through intelligent algorithms; then, all values ​​can be imported into the data template, so that when it is detected that there is no value corresponding to the key data type, the value corresponding to the key data type can be effectively and quickly calculated in combination with the set calculation method and all values; then, the second node can also be determined in combination with the initial organization name configured by the first node, so as to obtain the numerical deviation value of the key data type according to the first historical report corresponding to the first node and the second historical report corresponding to the second node, and the rationality of the generated data is verified through a cross-organization data verification method, and when the numerical deviation value is lower than a preset threshold, a target report is generated according to the data template, the value of the key data type and the first historical report, which not only improves the accuracy and completeness of the generated report, but also ensures the quality of the report uploaded to the direct reporting platform, thereby providing a complete and correct data basis for subsequent data applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A system architecture diagram of a cross-organization flexible data report generation method provided in an embodiment of the present application; Figure 2 An overall flow chart of a cross-organization flexible data report generation method provided in an embodiment of the present application; Figure 3 A schematic diagram of the effect of a data template provided in an embodiment of the present application; Figure 4 A schematic diagram of the effect of generating a report provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a cross-institutional flexible data report generation system provided in an embodiment of the present application; Figure 6A structural diagram of another cross-institutional flexible data report generation system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0018] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, although the embodiment may not be clearly recorded in the following text.

[0019] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.

[0020] See also Figure 1 , Figure 1 A system architecture diagram of a cross-institutional flexible data report generation method provided in an embodiment of the present application is shown.

[0021] like Figure 1As shown, the system architecture of the cross-institutional flexible data report generation method may at least include a report platform terminal, at least two node terminals and a direct reporting platform terminal. The report platform terminal may establish a connection with each node terminal respectively to process the report data uploaded by the third-party application installed on each node terminal to generate a corresponding report, and may upload the report to the corresponding direct reporting platform terminal according to the statistical survey requirements set by the node terminal (of course, it may also be displayed to the institution corresponding to the node terminal through the third-party application). For example, but not limited to, when the report data uploaded by any node terminal belongs to the data corresponding to the tax report, the report platform terminal may generate a tax report based on the report data, and upload the tax report to the corresponding tax statistics direct reporting platform terminal; or when the report data uploaded by any node terminal belongs to the data corresponding to the output value report, the report platform terminal may generate an output value report based on the report data, and upload the output value report to the corresponding output value statistics direct reporting platform terminal, and is not limited to this.

[0022] Among them, at least two node terminals may include, but are not limited to, node terminal A, node terminal B, and node terminal C. Node terminal A may be understood as a mobile terminal used by organization A to upload report data, node terminal B may be understood as a mobile terminal used by organization B to upload report data, and node terminal C may be understood as a mobile terminal used by organization C to upload report data, and in the embodiment of the present application, it may not be limited to the number of node terminals set. Here, before uploading report data to the report platform terminal, each node terminal also needs to register the organization on the report platform terminal, for example, but not limited to uploading the relevant information of the organization to the report platform terminal for review, so that the corresponding node terminal is registered after the review by the report platform terminal is correct, and before each node terminal uploads report data each time, the report platform terminal may also perform organization authentication, thereby avoiding abnormalities in the node terminal's organization login.

[0023] It can be understood that the report data uploaded by the node terminal can be but is not limited to the data corresponding to the tax report or the data corresponding to the output value report. The data corresponding to the tax report may specifically include the values ​​corresponding to at least one data type such as accounts receivable, inventory, original price of fixed assets, operating income, operating cost, operating profit, total assets and total liabilities; the data corresponding to the output value report may specifically include the values ​​corresponding to at least one data type such as product output, invoice quantity, invoice amount (excluding tax), sales unit price (excluding tax), output value, whether it is exported and whether it is a new product, etc., or the values ​​corresponding to at least one data type such as commissioned processing output, commissioned processing sales, processing fee income, whether it is commissioned by overseas enterprises and whether it is commissioned by domestic non-industrial enterprises, etc., or the values ​​corresponding to at least one data type such as contract unit price, calculated output value contract unit price, cumulative progress as of the end of last year, cumulative progress as of last month and cumulative progress as of this month. In addition, the file format of the report data can be but is not limited to a text file or an image file. The text file can be understood as a file containing text information or table information, and the image file can be understood as a file containing an invoice image or a bank receipt image, but is not limited to this.

[0024] It can also be understood that when the report platform terminal processes the report data uploaded by any node terminal (taking the first node as an example) to generate a corresponding report, it can specifically perform enterprise authentication on the first node based on the report generation request, and input the uploaded report data into the deep learning model after the authentication is passed, so as to more accurately extract multiple report data types and corresponding values ​​through intelligent algorithms; then, all values ​​can be imported into the data template, so that when it is detected that there is no value corresponding to the key data type, all values ​​are combined according to the algorithm set by the data template to effectively and quickly calculate the value corresponding to the key data type; then, the second node can also be determined in combination with the initial organization name configured by the first node, so as to obtain the value deviation value of the key data type according to the first historical report corresponding to the first node and the second historical report corresponding to the second node, and verify the rationality of the generated data through a cross-organization data verification method, and when the value deviation value is lower than the preset threshold, the target report is generated according to the data template, the value of the key data type and the first historical report, which not only improves the accuracy and completeness of the generated report, but also ensures the quality of the report uploaded to the direct reporting platform, thereby providing a complete and correct data basis for subsequent data applications.

[0025] See also Figure 2 , Figure 2 The overall flow chart of a cross-institutional flexible data report generation method provided in an embodiment of the present application is shown.

[0026] like Figure 2 As shown, the cross-institutional flexible data report generation method may at least include the following steps: Step 202: When a report generation request from a first node is received, authentication processing is performed on the first node based on the report generation request, and after determining that the first node has passed the authentication, the report data uploaded by the first node is obtained.

[0027] Specifically, when generating a report based on report data, the report platform terminal can, but is not limited to, when receiving a report generation request issued by any node terminal (i.e., the first node), perform institutional authentication processing on the node terminal based on the report generation request to determine whether the report generation request is issued by the organization configured by the node terminal, and after determining that it is issued by the organization configured by the node terminal (i.e., determining that the first node authentication is passed), a data reception request can be issued to the node terminal to obtain the report data uploaded by the node terminal. Here, the report generation request issued by the node terminal is generated by the user according to the information entered in the prompt input box after the user enters the institutional account information in the login interface of the third-party application installed by the node terminal. The input information can include, but is not limited to, the name of the organization corresponding to the node terminal and the data type to which the report data belongs. The data type to which the report data belongs can be the data corresponding to the tax report mentioned above or the data corresponding to the output value report.

[0028] It is understandable that after determining that the report generation request is not issued by the organization configured by the node terminal, the report platform terminal can, but is not limited to, issue an account warning message to the head of the organization based on the user information configured by the node terminal, so that the head of the organization can promptly change the password of the organization account or the login method of the organization account.

[0029] As an option of the embodiment of the present application, performing authentication processing on the first node based on the report generation request includes: Parsing the requesting organization name from the report generation request, and when detecting that the initial organization name configured by the first node is inconsistent with the requesting organization name, issuing a verification request according to the user information configured by the first node; When confirmation information of the verification request is received, determining that the first node is authenticated; or When it is detected that the initial organization name configured by the first node is consistent with the requesting organization name, it is determined that the first node is authenticated.

[0030] Specifically, in order to accurately determine whether the report generation request is issued by the organization configured by the node terminal, the report platform terminal can parse the organization name (i.e., the requesting organization name) input by the user from the report generation request after receiving the report generation request, and when it is detected that the organization name is inconsistent with the initial organization name configured by the node terminal, it indicates that there may be a login anomaly, and then the user information configured by the node terminal can be retrieved to issue a verification request based on the contact information in the user information. Here, the user information configured by the node terminal can include, but is not limited to, the person in charge of the organization and the corresponding contact information, and is not limited to this.

[0031] Next, when the report platform terminal receives the confirmation information corresponding to the verification request, it indicates that the report generation request is issued by the organization configured by the node terminal, and then it can be determined that the node terminal authentication is passed. At this time, the user can enter the data upload interface of the third-party application installed on the node terminal to select the corresponding upload method according to the file format of the report data to upload the report data.

[0032] It can be understood that when the report platform terminal does not receive the confirmation information corresponding to the verification request, it indicates that the report request is not issued by the organization configured by the node terminal. Then, the report platform terminal can send account warning information to the head of the organization based on the user information configured by the node terminal, so that the head of the organization can change the organization account password or the organization account login method in time.

[0033] In addition, when it is detected that the organization name is consistent with the initial organization name configured for the node terminal, it indicates that the report generation request is issued by the organization configured for the node terminal, and it can be determined that the node terminal authentication is passed. At this time, the user can enter the data upload interface of the third-party application installed on the node terminal to select the corresponding upload method according to the file format of the report data to upload the report data.

[0034] Step 204: When the report data is identified as an image file, the report data is input into a deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type.

[0035] Specifically, after receiving the report data uploaded by the node terminal, the report platform terminal can, but is not limited to, identify and process the file format of the report data, and when identifying that the report data is an image file of the type of invoice image or bank receipt image, the report data is input into the deep learning model, so as to more accurately extract valid multiple report data types and corresponding values ​​from the report data through intelligent algorithms. Here, taking the data type corresponding to the output value report as an example, when the report data includes the values ​​corresponding to the five data types of product output, invoiced quantity, invoiced amount (excluding tax), whether it is exported, and whether it is a new product, the deep learning model can be used to extract the values ​​corresponding to the three data types of product output, invoiced quantity, and invoiced amount (excluding tax) from the report data, and is not limited to this.

[0036] It can be understood that the deep learning model mentioned in the embodiment of the present application can be trained by multiple report sample images and report data labels manually marked according to each report sample image. The report data labels may include multiple standard data types and numerical values ​​corresponding to each standard data type. For example, taking the report sample image as an invoice image mainly based on products as an example, product output, invoice quantity, invoice amount (excluding tax), sales unit price and product output value can be used as standard data types, but are not limited to this.

[0037] As another option of the embodiment of the present application, before inputting the report data into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, after identifying that the report data is an image file, the method further includes: Performing image enhancement processing on the report data, and marking characteristic pattern contours in the report data after image enhancement processing; Identify at least two edge pixel values ​​of the characteristic pattern contour, and determine whether the characteristic pattern contour has edge anomalies based on all edge pixel values; The report data is input into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, including: When it is determined that there is no edge anomaly in the characteristic pattern contour, the report data is input into the deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type.

[0038] When the report data is in the form of an image file, in order to ensure the authenticity and reliability of the report data, the report data may also be processed by image recognition to determine whether the report data has been modified or fabricated by judging whether there are edge anomalies in the feature pattern.

[0039] Specifically, after identifying that the report data is an image file, the report platform terminal may also perform image enhancement processing on the report image corresponding to the report data, such as but not limited to grayscale transformation processing, spatial domain filtering processing, or color space change processing, to improve the visual effect of the report image and improve the image quality of the report data, and identify multiple pattern contours in the processed report image, so as to mark one or more characteristic pattern contours in the processed report image according to the multiple pattern contours. Here, the characteristic pattern contour can be understood as a pattern contour that is easily modified or fabricated in the report image, such as but not limited to a rectangular contour frame or a circular contour frame containing a seal (physical seal or electronic seal), and a rectangular contour frame containing a signature (physical signature or electronic signature). Of course, in the embodiment of the present application, a rectangular contour frame containing numerical values ​​corresponding to various data types can also be used as a characteristic pattern contour, but it is not limited to this.

[0040] Next, after marking the characteristic pattern outline, the reporting platform terminal can also determine the edge area around the characteristic pattern outline, for example but not limited to taking all areas within a preset distance threshold from the characteristic pattern outline as edge areas (that is, including partial areas inside the characteristic pattern outline and partial areas outside the characteristic pattern outline), and performing identification processing on the edge area through a pixel recognition algorithm to obtain pixel values ​​corresponding to multiple positions in the edge area. Here, the pixel recognition algorithm can be an image processing method well known in the art, and of course it can also be replaced by a preset control program, but it will not be described in detail here.

[0041] It can be understood that when a large difference is detected between any two adjacent pixel values ​​among all pixel values ​​(for example, the difference between at least one color channel exceeds a preset color difference), or a large pixel change rate is detected between any three consecutive pixel values ​​(for example, the ratio of the difference between a certain color channel corresponding to the first two adjacent pixel values ​​and the difference between the same color channel corresponding to the last two adjacent pixel values ​​exceeds a preset difference ratio), it can be determined that there is an edge anomaly in the characteristic pattern contour, that is, the report data may be modified or fabricated, and the report platform terminal can send data anomaly information to the head of the organization based on the user information configured by the node terminal, so that the head of the organization can check the uploaded report data in a timely manner.

[0042] When it is detected that the difference between any two adjacent pixel values ​​among all pixel values ​​is relatively small, or when it is detected that the pixel change rate between any three consecutive pixel values ​​is small, it can be determined that there is no edge anomaly in the characteristic pattern contour, that is, the report data has not been modified or fabricated, and the report platform terminal can then input the report data into the deep learning model to more accurately extract valid multiple report data types and corresponding values ​​from the report data through intelligent algorithms.

[0043] Of course, in the embodiment of the present application, it is also possible but not limited to combining curve tools or image analysis tools to identify and process the pixel values ​​corresponding to multiple positions in the edge area to analyze the changes in all pixel values, and then determine whether there are edge anomalies in the characteristic pattern contour, which will not be elaborated here.

[0044] As another option of the embodiment of the present application, after obtaining the report data uploaded by the first node, the method further includes: When the report data is identified as a text file, the report data is standardized; Based on a preset data type library, the standardized report data is matched to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the preset data type library includes at least two standard data types and a synonymous data type corresponding to each standard data type.

[0045] Specifically, when it is identified that the file format of the report data uploaded by the node terminal is a text file, in order to more quickly and accurately extract effective multiple report data types and corresponding values ​​from the report data, the report platform terminal can also, but not limited to, perform standardization processing on the report text corresponding to the report data, such as, but not limited to, case conversion processing, character conversion processing, or symbol removal processing, so as to improve the processing efficiency of the report text corresponding to the report data, and can perform matching processing on the processed report text based on the preset data type library to extract the matched multiple report data types and corresponding values. Here, the preset data type library can include at least two standard data types and synonymous data types corresponding to each standard data type, such as, but not limited to, when a certain standard data type is a sales unit price, the corresponding synonymous data type can be a sales price, a unit price, and a selling price, so as to detect whether there is a phrase consistent with the sales unit price, the sales price, the unit price, or the selling price in the processed report text, and when it is detected that there is a phrase consistent with the sales unit price, the sales price, the unit price, or the selling price, the corresponding value is extracted as the value corresponding to the sales unit price, and is not limited to this.

[0046] Step 206: Import the values ​​corresponding to all report data types into the data template, and when it is detected that the value corresponding to the key data type does not exist in the imported data template, calculate the value corresponding to the key data type based on the values ​​corresponding to all report data types.

[0047] Specifically, after extracting a variety of valid report data types and corresponding values ​​from the report data, the report platform terminal can, but is not limited to, import the values ​​corresponding to all report data types into a data template (the data template can be classified according to all report data types and values ​​corresponding to different product names) to quickly determine whether there are values ​​corresponding to key data types through the data template after the data is imported. It can be understood that, taking the data type corresponding to the output value report as an example, the key data type can be product output value, processing output value or progress output value, and when it is detected that the value corresponding to the key data type does not exist in the data template after the data is imported, the values ​​corresponding to at least two report data types can be substituted into the preset calculation formula to calculate the value corresponding to the key data type.

[0048] For example, but not limited to, taking the key data type as processing output value, when the report platform terminal identifies processing fee income, commissioned processing sales volume, and commissioned processing output in all report data types, the numerical values ​​corresponding to the processing fee income, the numerical values ​​corresponding to the commissioned processing sales volume, and the numerical values ​​corresponding to the commissioned processing output can be substituted into the preset processing output value formula to calculate the numerical value corresponding to the processing output value. Here, the preset processing output value formula can be, but is not limited to, as follows: Processing output value = processing fee income / commissioned processing sales volume * commissioned processing output For another example, but not limited to, taking the key data type as progress output value, when the report platform terminal identifies the cumulative progress as of this month, the cumulative progress as of last month, and the calculated output value contract unit price among all report data types, the numerical value corresponding to the cumulative progress as of this month, the numerical value corresponding to the cumulative progress as of last month, and the numerical value corresponding to the calculated output value contract unit price can be substituted into the preset progress output value formula to calculate the numerical value corresponding to the progress output value. Here, the preset progress output value formula can be, but is not limited to, as follows: Progress output value = (this month's cumulative progress - last month's cumulative progress) * calculated output value contract unit price As another option in the embodiment of the present application, the key data type is product output value; Calculate the values ​​corresponding to the key data types based on the values ​​corresponding to all report data types, including: When all report data types include product output, invoiced quantity and invoiced amount, the numerical value corresponding to the product output, the numerical value corresponding to the invoiced quantity and the numerical value corresponding to the invoiced amount are substituted into the preset first formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output and sales unit price, the numerical value corresponding to the product output and the numerical value corresponding to the sales unit price are substituted into the preset second formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output, determine the value corresponding to the invoiced quantity of at least one sub-product and the value corresponding to the invoiced amount of each sub-product according to the product name corresponding to the product output; The values ​​corresponding to the invoiced quantities of all sub-products are summed to obtain a first summation result, and the values ​​corresponding to the invoiced amounts of all sub-products are summed to obtain a second summation result; The first summation result, the second summation result and the product output are substituted into a preset third formula to calculate a numerical value corresponding to the product output value.

[0049] Specifically, taking the key data type as product output value as an example, when the report platform terminal identifies the product output, invoiced quantity and invoiced amount corresponding to the same product name in all report data types, the numerical value corresponding to the product output, the numerical value corresponding to the invoiced quantity and the numerical value corresponding to the invoiced amount can be substituted into the preset first formula to calculate the numerical value corresponding to the product output value. Here, the preset first formula can be but is not limited to the following: Product output value = invoice amount / invoice quantity * product output It should be noted that no matter whether there is a sales unit price in all report data types, the value corresponding to the product output value is calculated in combination with the above-mentioned preset first formula.

[0050] It is understandable that when the report platform terminal identifies the product output and sales unit price corresponding to the same product name in all report data types, the numerical value corresponding to the product output and the numerical value corresponding to the sales unit price can be substituted into the preset second formula to calculate the numerical value corresponding to the product output value. Here, the preset second formula can be but is not limited to the following: Product output value = product output * sales unit price It is understandable that when the report platform terminal only identifies the product output corresponding to the same product name in all report data types, it can also determine all corresponding sub-product names based on the product name, so as to query the numerical value corresponding to the invoiced quantity of each sub-product name and the numerical value corresponding to the invoiced amount of each sub-product in the data template after the data is imported. It is understandable that when the report platform terminal determines all corresponding sub-product names based on the product name, it can, but is not limited to, query all sub-product names corresponding to the product name with reference to a preset product classification list, and the preset product classification list can include at least two product names and all sub-product names corresponding to each product name, and is not limited to this.

[0051] Next, the values ​​corresponding to the invoiced quantities of all sub-products may be summed to obtain a first summed result, and the values ​​corresponding to the invoiced amounts of all sub-products may be summed to obtain a second summed result, and the first summed result, the second summed result, and the product output may be substituted into a preset third formula to calculate the value corresponding to the product output value. Here, the preset third formula may be, but is not limited to, as follows: Product output value = second summation result / first summation result * product output Also see here Figure 3 The effect diagram of a data template provided by an embodiment of the present application is shown as follows: Figure 3 As shown, the output value template after importing data includes five product names, which can be represented as product A, product B, product C, product c1 and product c2 (product c1 and product c2 are sub-product names of product C). Among them, there is no value corresponding to the product output value (that is, the value corresponding to the key data type) for product A, product B and product C. Since product A corresponds to output A, quantity A and amount A, the value corresponding to the product output value can be calculated by the first preset formula mentioned above; since product B corresponds to output B and unit price B, the value corresponding to the product output value can be calculated by the second preset formula mentioned above; since product C corresponds to output C, but the sub-product c1 of product C corresponds to quantity c1 and amount c2, and the sub-product c2 corresponds to quantity c2 and amount c2, the value corresponding to the product output value can be calculated by the third preset formula mentioned above.

[0052] Step 208: Determine the second node according to the initial organization name configured for the first node, and obtain the first historical report corresponding to the first node and the second historical report corresponding to the second node.

[0053] Specifically, after calculating the numerical value corresponding to the key data type, in order to provide a data basis for the cross-institutional data verification method and ensure the reliability of the verification method, the reporting platform terminal can, but is not limited to, determine the institution type based on the initial institution name configured by the node terminal, so as to screen out at least one associated node terminal corresponding to the institution type from the institution names configured by all other node terminals, that is, the second node, and can extract all historical reports generated within a specified time period under the report storage path corresponding to the node terminal, that is, the first historical report, and extract all historical reports generated within the same time period under the report storage path corresponding to the associated node terminal, that is, the second historical report.

[0054] It can be understood that the first historical report may include one or more historical reports generated by the node terminal within a specified time period, and the second historical report may include one or more historical reports generated by the associated node terminal within the same time period. The historical report may at least include the report name, the unified social credit code of the organization, the organization name, the form number, the validity period, the integrated data, the organization head and the contact information. The integrated data can be obtained by the data template statistics after the data is imported, and is not limited to this.

[0055] Step 210: determine the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report, and when the numerical deviation value is lower than a preset threshold, generate a target report based on the imported processed data template, the numerical value corresponding to the key data type and the first historical report.

[0056] Specifically, after determining the first historical report and the second historical report, the report platform terminal can, but is not limited to, extract all historical values ​​corresponding to the key data type in the first historical report, so as to generate a value change curve corresponding to the key data type at the node terminal based on all historical values ​​and the values ​​corresponding to the key data type; in addition, the report platform terminal can also extract all historical values ​​corresponding to the key data type in the second historical report, so as to generate a value change curve corresponding to the key data type at the associated node terminal based on all historical values, and use the similarity between the two value change curves as the value deviation value. Here, the value change curve can be understood as a smooth curve of the value corresponding to the key data type changing over time. When calculating the similarity between the two value change curves, partial change curves in the same time period can be intercepted from the two value change curves to calculate the similarity of the two partial change curves in the same time period, and the similarity calculation method in the embodiment of the present application is a technical means well known in the art, but will not be elaborated here.

[0057] Further, after obtaining the numerical deviation value, when the report platform terminal detects that the numerical deviation value is lower than the preset threshold value, it indicates that the numerical value corresponding to the key data type is reliable, and then the numerical value corresponding to the key data type can be imported into the processed data template, and a variety of data types and corresponding numerical values ​​that can be displayed on the report are extracted from the data simulation, and the final target report is generated by integrating the extracted multiple data types and corresponding numerical values, the historical numerical values ​​corresponding to the key data types in the first historical report, and the organization information and user information corresponding to the node terminal. Here, taking the key data type as product output value as an example, the multiple data types that can be displayed on the report extracted from the data simulation can include but are not limited to product name, product output, and product difference, and after extracting the historical numerical value corresponding to the product output value in the first historical report, the report platform terminal can also sum the historical numerical value and the numerical value corresponding to the product output value to obtain the cumulative product difference up to this month, and the cumulative product difference can be displayed on the target report.

[0058] It can be understood that when the reporting platform terminal detects that the numerical deviation value is higher than the preset threshold, it indicates that there is an abnormality in the numerical value corresponding to the key data type, which may be caused by inaccurate report data uploaded by the node terminal. Then, based on the user information configured by the node terminal, data abnormality information can be sent to the head of the organization so that the head of the organization can check the uploaded report data in time.

[0059] As another option of the embodiment of the present application, determining the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report includes: At a preset time interval, at least two first values ​​corresponding to the product output value are selected from the first historical report, and a fitting process is performed on all the first values ​​and the values ​​corresponding to the product output value to obtain a first fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent first moments and the numerical value corresponding to each first moment are determined in the first fitting curve, and the first numerical value change value is calculated according to all the first moments and the numerical value corresponding to each first moment; At a preset time interval, at least two second values ​​corresponding to the product output value are selected from the second historical report, and a fitting process is performed on all the second values ​​to obtain a second fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent second moments and the numerical value corresponding to each second moment are determined in the second fitting curve, and the second numerical value change value is calculated according to all the second moments and the numerical value corresponding to each second moment; The difference between the first numerical change value and the second numerical change value is calculated to obtain a numerical deviation value corresponding to the key data type.

[0060] Specifically, taking the product output value as an example of a key data type, when the reporting platform terminal obtains the numerical deviation value of the product difference, it can first filter out at least two historical values ​​(that is, the first value) corresponding to the product output value from the first historical report according to a preset time interval. For example, when the preset time interval is six months, and the first historical report includes historical reports corresponding to each month, the historical values ​​corresponding to the product output value can be filtered out from the historical reports generated in the last six months, and all historical values ​​and the time corresponding to each historical value can be fitted based on the least squares algorithm to obtain a first fitting curve with higher accuracy.

[0061] Then, after obtaining the first fitting curve, the reporting platform terminal can also determine two first moments adjacent to the moment and the numerical value corresponding to each first moment in the first fitting curve according to the moment of the numerical value corresponding to the product output value, and process the two first moments and the corresponding numerical values ​​through slope calculation to obtain a first numerical change value that can characterize the numerical change trend corresponding to the product output value.

[0062] In the same way, the reporting platform terminal can also determine the corresponding second numerical change value based on the second historical report, which will not be elaborated here, and the difference between the first numerical change value and the second numerical change value can be used as the numerical deviation value corresponding to the product output value.

[0063] As another option of the embodiment of the present application, generating a target report according to the imported processed data template, the value corresponding to the key data type and the first historical report includes: Summing all the first values ​​and the values ​​corresponding to the product output values ​​to obtain a third summation result; Generate a target report based on the imported processed data template, the numerical value corresponding to the product output value and the third summation result.

[0064] Specifically, taking product output value as an example of a key data type, in order to present a more intuitive annual numerical growth effect to the organization, the report platform terminal can also filter out all historical values ​​corresponding to January of this year to this month (that is, the first value) in the first historical report, and sum all historical values ​​and values ​​corresponding to the product output value to obtain the value of the product output value accumulated from this year to this month (that is, the third summation result), and can extract a variety of data types and corresponding values ​​that can be displayed on the report from the data simulation, and generate the final target report by integrating the extracted multiple data types and corresponding values, the value of the product output value accumulated from this year to this month, and the organization information and user information corresponding to the node terminal.

[0065] Also see here Figure 4 The schematic diagram of the effect of generating a report provided by the embodiment of the present application is shown as follows: Figure 4 As shown, taking the name of the generated report as the total value of production and sales and output production report as an example, the report may include the name of the organization corresponding to the node terminal, the unified social credit code, the person in charge of the organization and the contact information, and may also include the product name, product code, unit of measurement, product output and product output value of this month corresponding to multiple products extracted from the data template, and may also include the product output value of 1-this month corresponding to multiple products, and the product output value of 1-this month may be the sum of the product output value of the corresponding product this month and all historical product output values ​​from January to this month. In addition, the report may also include the table number and validity period generated according to the preset control program, and is not limited to this.

[0066] See also Figure 5 , Figure 5 A structural schematic diagram of a cross-institutional flexible data report generation system provided in an embodiment of the present application is shown.

[0067] like Figure 5 As shown, the cross-institutional flexible data report generation system may include at least a node authentication module 501, a data extraction module 502, a numerical calculation module 503, a data query module 504 and a report generation module 505, wherein: The node authentication module 501 is used for, when receiving a report generation request from a first node, performing authentication processing on the first node based on the report generation request, and obtaining the report data uploaded by the first node after determining that the first node has passed the authentication; The data extraction module 502 is used to input the report data into the deep learning model when the report data is identified as an image file, and obtain at least two report data types and a numerical value corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and a report data label marked by each report sample image, and each report data label includes at least two standard data types and a numerical value corresponding to each standard data type; The numerical calculation module 503 is used to import the numerical values ​​corresponding to all report data types into the data template, and when it is detected that the numerical value corresponding to the key data type does not exist in the data template after the import processing, calculate the numerical value corresponding to the key data type according to the numerical values ​​corresponding to all report data types; The data query module 504 is used to determine the second node according to the initial organization name configured by the first node, and obtain the first historical report corresponding to the first node and the second historical report corresponding to the second node; The report generation module 505 is used to determine the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report, and when the numerical deviation value is lower than a preset threshold, generate a target report according to the imported processed data template, the numerical value corresponding to the key data type and the first historical report.

[0068] In some possible embodiments, performing authentication processing on the first node based on the report generation request includes: Parsing the requesting organization name from the report generation request, and when detecting that the initial organization name configured by the first node is inconsistent with the requesting organization name, issuing a verification request according to the user information configured by the first node; When confirmation information of the verification request is received, determining that the first node is authenticated; or When it is detected that the initial organization name configured by the first node is consistent with the requesting organization name, it is determined that the first node is authenticated.

[0069] In some possible embodiments, before inputting the report data into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, after identifying that the report data is an image file, the method further includes: Performing image enhancement processing on the report data, and marking characteristic pattern contours in the report data after image enhancement processing; Identify at least two edge pixel values ​​of the characteristic pattern contour, and determine whether the characteristic pattern contour has edge anomalies based on all edge pixel values; The report data is input into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, including: When it is determined that there is no edge anomaly in the characteristic pattern contour, the report data is input into the deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type.

[0070] In some possible embodiments, after obtaining the report data uploaded by the first node, the method further includes: When the report data is identified as a text file, the report data is standardized; Based on a preset data type library, the standardized report data is matched to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the preset data type library includes at least two standard data types and a synonymous data type corresponding to each standard data type.

[0071] In some possible embodiments, the key data type is product output value; Calculate the values ​​corresponding to the key data types based on the values ​​corresponding to all report data types, including: When all report data types include product output, invoiced quantity and invoiced amount, the numerical value corresponding to the product output, the numerical value corresponding to the invoiced quantity and the numerical value corresponding to the invoiced amount are substituted into the preset first formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output and sales unit price, the numerical value corresponding to the product output and the numerical value corresponding to the sales unit price are substituted into the preset second formula to calculate the numerical value corresponding to the product output value; or When all report data types include product output, determine the value corresponding to the invoiced quantity of at least one sub-product and the value corresponding to the invoiced amount of each sub-product according to the product name corresponding to the product output; The values ​​corresponding to the invoiced quantities of all sub-products are summed to obtain a first summation result, and the values ​​corresponding to the invoiced amounts of all sub-products are summed to obtain a second summation result; The first summation result, the second summation result and the product output are substituted into a preset third formula to calculate a numerical value corresponding to the product output value.

[0072] In some possible embodiments, determining the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report includes: At a preset time interval, at least two first values ​​corresponding to the product output value are selected from the first historical report, and a fitting process is performed on all the first values ​​and the values ​​corresponding to the product output value to obtain a first fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent first moments and the numerical value corresponding to each first moment are determined in the first fitting curve, and the first numerical value change value is calculated according to all the first moments and the numerical value corresponding to each first moment; At a preset time interval, at least two second values ​​corresponding to the product output value are selected from the second historical report, and a fitting process is performed on all the second values ​​to obtain a second fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent second moments and the numerical value corresponding to each second moment are determined in the second fitting curve, and the second numerical value change value is calculated according to all the second moments and the numerical value corresponding to each second moment; The difference between the first numerical change value and the second numerical change value is calculated to obtain a numerical deviation value corresponding to the key data type.

[0073] In some possible embodiments, generating a target report according to the imported processed data template, the values ​​corresponding to the key data types, and the first historical report includes: Summing all the first values ​​and the values ​​corresponding to the product output values ​​to obtain a third summation result; Generate a target report based on the imported processed data template, the numerical value corresponding to the product output value and the third summation result.

[0074] See also Figure 6 , Figure 6 A structural schematic diagram of another cross-institutional flexible data report generation system provided in an embodiment of the present application is shown.

[0075] like Figure 6 As shown, the cross-institution flexible data report generation system 600 may include at least one processor 601 , at least one network interface 604 , a user interface 603 , a memory 605 , and at least one communication bus 602 .

[0076] The communication bus 602 may be used to realize the connection and communication among the above-mentioned components.

[0077] The user interface 603 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0078] The network interface 604 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0079] Among them, the processor 601 may include one or more processing cores. The processor 601 uses various interfaces and lines to connect the various parts within the cross-institutional flexible data report generation system 600, and executes various functions and processes data of the cross-institutional flexible data report generation system 600 by running or executing instructions, programs, code sets or instruction sets stored in the memory 605, and calling data stored in the memory 605. Optionally, the processor 601 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 601 can integrate one or a combination of CPU, GPU, modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application program; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 601, and it can be implemented by a single chip.

[0080] The memory 605 may include a RAM or a ROM. Optionally, the memory 605 includes a non-transitory computer-readable medium. The memory 605 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 605 may optionally be at least one storage device located away from the aforementioned processor 601. As Figure 6 As shown, the memory 605 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an inter-institutional flexible data report generation application.

[0081] Specifically, the processor 601 may be used to call the cross-organization flexible data report generation application stored in the memory 605, and specifically perform the following operations: When receiving a report generation request from the first node, performing authentication processing on the first node based on the report generation request, and after determining that the first node passes the authentication, obtaining the report data uploaded by the first node; When the report data is identified as an image file, the report data is input into a deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and report data labels marked by each report sample image, and each report data label includes at least two standard data types and a numerical value corresponding to each standard data type; Import the values ​​corresponding to all report data types into the data template, and when it is detected that the value corresponding to the key data type does not exist in the data template after the import processing, calculate the value corresponding to the key data type based on the values ​​corresponding to all report data types; Determine the second node according to the initial organization name configured by the first node, and obtain the first historical report corresponding to the first node and the second historical report corresponding to the second node; Based on the first historical report and the second historical report, a numerical deviation value corresponding to the key data type is determined, and when the numerical deviation value is lower than a preset threshold, a target report is generated based on the imported processed data template, the numerical value corresponding to the key data type and the first historical report.

[0082] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0083] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0084] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0085] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0086] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

Claims

1. A cross-institutional flexible data report generation method, characterized in that: include: When receiving a report generation request from a first node, performing authentication processing on the first node based on the report generation request, and after determining that the first node passes the authentication, obtaining report data uploaded by the first node; When the report data is identified as an image file, the report data is input into a deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and report data labels marked by each of the report sample images, and each of the report data labels includes at least two standard data types and a numerical value corresponding to each of the standard data types; Importing all the numerical values ​​corresponding to the report data types into the data template, and when detecting that the numerical value corresponding to the key data type does not exist in the data template after the import processing, calculating the numerical value corresponding to the key data type according to the numerical values ​​corresponding to all the report data types; Determine a second node according to the initial organization name configured by the first node, and obtain a first historical report corresponding to the first node and a second historical report corresponding to the second node; Based on the first historical report and the second historical report, a numerical deviation value corresponding to the key data type is determined, and when the numerical deviation value is lower than a preset threshold, a target report is generated based on the imported data template, the numerical value corresponding to the key data type, and the first historical report.

2. The method according to claim 1, characterized in that The performing authentication processing on the first node based on the report generation request includes: Parsing the requesting organization name from the report generation request, and when detecting that the initial organization name configured by the first node is inconsistent with the requesting organization name, issuing a verification request according to the user information configured by the first node; When confirmation information of the verification request is received, determining that the first node is authenticated; or When it is detected that the initial organization name configured by the first node is consistent with the requesting organization name, it is determined that the first node is authenticated.

3. The method according to claim 1, characterized in that: Before inputting the report data into the deep learning model to obtain at least two report data types and numerical values ​​corresponding to each report data type, after identifying that the report data is an image file, the method further includes: Performing image enhancement processing on the report data, and marking a characteristic pattern outline in the report data after the image enhancement processing; Identify at least two edge pixel values ​​of the characteristic pattern contour, and determine whether the characteristic pattern contour has edge anomalies according to all the edge pixel values; The step of inputting the report data into a deep learning model to obtain at least two report data types and a value corresponding to each report data type includes: When it is determined that there is no edge anomaly in the characteristic pattern contour, the report data is input into a deep learning model to obtain at least two report data types and a numerical value corresponding to each report data type.

4. The method according to claim 1, characterized in that: After obtaining the report data uploaded by the first node, the method further includes: When the report data is identified as a text file, standardizing the report data; The standardized report data is matched based on a preset data type library to obtain at least two report data types and a numerical value corresponding to each report data type; wherein the preset data type library includes at least two standard data types and a synonymous data type corresponding to each of the standard data types.

5. The method according to claim 1, characterized in that The key data type is product output value; The calculating the value corresponding to the key data type according to the values ​​corresponding to all the report data types includes: When all the report data types include product output, invoiced quantity and invoiced amount, the numerical value corresponding to the product output, the numerical value corresponding to the invoiced quantity and the numerical value corresponding to the invoiced amount are substituted into a preset first formula to calculate the numerical value corresponding to the product output value; or When all the report data types include the product output and the sales unit price, the numerical value corresponding to the product output and the numerical value corresponding to the sales unit price are substituted into the preset second formula to calculate the numerical value corresponding to the product output value; or When all the report data types include the product output, determining the value corresponding to the invoiced quantity of at least one sub-product and the value corresponding to the invoiced amount of each sub-product according to the product name corresponding to the product output; The numerical values ​​corresponding to the invoiced quantities of all the sub-products are summed to obtain a first summed result, and the numerical values ​​corresponding to the invoiced amounts of all the sub-products are summed to obtain a second summed result; The first summation result, the second summation result and the product output are substituted into a preset third formula to calculate a numerical value corresponding to the product output value.

6. The method according to claim 5, characterized in that The determining, based on the first historical report and the second historical report, a numerical deviation value corresponding to the key data type includes: At a preset time interval, at least two first values ​​corresponding to the product output value are selected from the first historical report, and a fitting process is performed on all the first values ​​and the values ​​corresponding to the product output value to obtain a first fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent first moments and the numerical value corresponding to each first moment are determined in the first fitting curve, and a first numerical value change value is calculated according to all the first moments and the numerical value corresponding to each first moment; According to the preset time interval, at least two second values ​​corresponding to the product output value are screened out from the second historical report, and a fitting process is performed on all the second values ​​to obtain a second fitting curve; According to the moment of the numerical value corresponding to the product output value, two adjacent second moments and the numerical value corresponding to each second moment are determined in the second fitting curve, and a second numerical value change value is calculated according to all the second moments and the numerical value corresponding to each second moment; A difference calculation is performed on the first numerical change value and the second numerical change value to obtain a numerical deviation value corresponding to the key data type.

7. The method according to claim 6, characterized in that The step of generating a target report according to the imported data template, the numerical value corresponding to the key data type, and the first historical report includes: Summing all the first values ​​and the values ​​corresponding to the product output values ​​to obtain a third summation result; A target report is generated according to the imported data template, the numerical value corresponding to the product output value and the third summation result.

8. A cross-institutional flexible data report generation system, characterized in that: include: A node authentication module, configured to, when receiving a report generation request from a first node, perform authentication processing on the first node based on the report generation request, and after determining that the first node has passed the authentication, obtain the report data uploaded by the first node; A data extraction module, for inputting the report data into a deep learning model when identifying the report data as an image file, to obtain at least two report data types and numerical values ​​corresponding to each report data type; wherein the deep learning model is trained by a plurality of report sample images and report data labels marked by each of the report sample images, and each of the report data labels includes at least two standard data types and numerical values ​​corresponding to each of the standard data types; A numerical calculation module, used for importing the numerical values ​​corresponding to all the report data types into the data template, and when detecting that the numerical value corresponding to the key data type does not exist in the data template after the import processing, calculating the numerical value corresponding to the key data type according to the numerical values ​​corresponding to all the report data types; A data query module, used to determine the second node according to the initial organization name configured by the first node, and obtain a first historical report corresponding to the first node and a second historical report corresponding to the second node; A report generation module is used to determine the numerical deviation value corresponding to the key data type based on the first historical report and the second historical report, and when the numerical deviation value is lower than a preset threshold, generate a target report based on the imported data template, the numerical value corresponding to the key data type and the first historical report.

9. A cross-institutional flexible data report generation system, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the steps of the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Chinese character recognition method and device for financial statements

    CN108491789A

  • Data processing method based on information identification technology and related device

    CN110264288A

  • Financial statement automatic identification method based on deep learning framework

    CN111814598A

  • Report analysis method and device

    CN112445849A

  • Data report processing method and device, computer equipment and storage medium

    CN114896955A