Data Screening Method for the Additional Deduction System

Abstract

The present invention provides a data screening method for a additional deduction system, which relates to data processing technology and includes: configuring a project management module for a block node to store a first project table of R & D behaviors, and extracting and processing the first project table to obtain details of project related parties and details of R & D expenditures; configuring a project inquiry module for a block node to store a second project table of R & D service revenues, and extracting and processing the second project table to obtain details of auxiliary R & D; if the block server determines that a data screening requirement for additional deduction is uploaded by a first block node, it determines the projects to be screened based on the current time point and the data axis corresponding to the first project table, and conducts data screening based on the block database of the first block node and the second project table of the second block node determined by the projects to be screened; the block server statistically generates a corresponding total screening result for the data screening result, and obtains the compliant data for additional deduction based on the total screening result and broadcasts it.

Description

Technical Field

[0001] The present invention relates to data processing technologies, and in particular, to a data screening method for a system of additional deductions for R & D expenses. Background Art

[0002] In the current financial management and tax declaration work of enterprises, the additional deductions for R & D expenses are an extremely important and complex business. When enterprises carry out R & D activities, a large amount of data related to R & D projects will be generated, including details of R & D expenditures, information on project related parties, data on R & D service revenues, etc. Accurately and efficiently managing and screening this data is of crucial significance to enterprises. However, with the increase in the number of R & D projects of enterprises, the improvement of business complexity, and the continuous refinement of tax policies, traditional data management and screening methods are facing huge challenges.

[0003] When the prior art processes data screening for the system of additional deductions for R & D expenses, it mostly adopts a relatively scattered and isolated method. For example, when conducting data screening, it mostly relies on manual comparison and screening. This method is not only inefficient, prone to human errors, but also difficult to comprehensively and accurately analyze large-scale and complex data.

[0004] Therefore, how to automatically process associated data in combination with data screening requirements to improve efficiency and accuracy has become an urgent problem to be solved. Summary of the Invention

[0005] An embodiment of the present invention provides a data screening method for a system of additional deductions for R & D expenses, which can automatically process associated data in combination with data screening requirements to improve efficiency and accuracy.

[0006] In a first aspect of an embodiment of the present invention, there is provided a data screening method for a system of additional deductions for R & D expenses, including:

[0007] Configuring a project management module for a block node to store a first project table of R & D behaviors, and extracting and processing the first project table to obtain details of project related parties and details of R & D expenditures;

[0008] Configuring a project inquiry module for a block node to store a second project table of R & D service revenues, and extracting and processing the second project table to obtain details of auxiliary R & D;

[0009] If the block server determines that a data screening requirement for additional deductions is uploaded by a first block node, it determines projects to be screened based on the current time point and the data axis corresponding to the first project table, and conducts data screening based on the block database of the first block node and the second project table of the second block node for the projects to be screened;

[0010] The block server generates corresponding total screening results based on the data screening results, and obtains the compliance data for additional deductions based on the total screening results and broadcasts it.

[0011] Optionally, a project management module is configured for the block node, which is used to store the first project table of the R & D behavior, and extract and process the first project table to obtain the project related party details and R & D expenditure details, including:

[0012] When the block node determines that the management end has an upload requirement, it generates an initial project table and feeds it back to the management end, and configures all expenditure data and project requirement data for the initial project table. The expenditure data includes multiple expenditure information with text and / or images;

[0013] Based on the project requirement data, a project timeline and a project cost axis corresponding to the project table are established. The cut-off point of the project timeline is the cut-off time, and the cut-off point of the project cost axis is the project budget;

[0014] Perform text and / or image extraction and processing on the first project table, obtain the corresponding project related party details and R & D expenditure details, fill them into the project timeline and the project cost axis, and perform segmentation processing.

[0015] Optionally, the performing text and / or image extraction and processing on the first project table, obtaining the corresponding project related party details and R & D expenditure details, filling them into the project timeline and the project cost axis, and performing segmentation processing includes:

[0016] Perform text and / or image extraction and processing on the first project table to determine the project related parties, R & D expenditures, and related times of each sub-project. Each sub-project has corresponding text and / or images;

[0017] Based on the related time, establish time nodes corresponding to each sub-project on the project timeline, and store the corresponding text and / or images and project related parties in the first project table corresponding to the time nodes;

[0018] Based on the order of the project related parties on the project timeline, establish stepped cost nodes corresponding to each sub-project for the R & D expenditures of each project related party on the project cost axis, and store the corresponding text and / or images and project related parties in the first project table corresponding to the stepped cost nodes.

[0019] Optionally, the based on the order of the project related parties on the project timeline, establishing stepped cost nodes corresponding to each sub-project for the R & D expenditures of each project related party on the project cost axis, and storing the corresponding text and / or images and project related parties in the first project table corresponding to the stepped cost nodes includes:

[0020] The unit amount corresponding to each pixel point is obtained through the project budget calculation process of the pixel point length and cut-off point based on the project cost axis;

[0021] Each project-related party and the corresponding R & D expenditure are sequentially extracted in the order of the project-related parties on the project time axis, and the number of pixel points corresponding to the project-related party is calculated based on the R & D expenditure and the unit amount;

[0022] Based on the number of pixel points, the stepped cost nodes corresponding to each sub-project are sequentially determined, and a storage space corresponding to each stepped cost node is established to store the corresponding text and / or image and project-related party.

[0023] Optionally, the stepped cost nodes corresponding to each sub-project are sequentially determined based on the number of pixel points, and a storage space corresponding to each stepped cost node is established to store the corresponding text and / or image and project-related party, including:

[0024] If it is determined that there are no stepped cost nodes corresponding to other sub-projects in the front part of the project cost axis, the first point of the project cost axis is used as the starting point, and the stepped cost nodes of this time are sequentially selected and obtained according to the number of pixel points;

[0025] If it is determined that there are stepped cost nodes corresponding to other sub-projects in the front part of the project cost axis, the last point of the last other sub-project in the project cost axis is used as the starting point, and the stepped cost nodes of this time are sequentially selected and obtained according to the number of pixel points;

[0026] The proportion of the number of pixel points of each sub-project in the total number of pixel points is calculated and stored in the corresponding storage space.

[0027] Optionally, a project inquiry module is configured for the block node, which is used to store the second project table of R & D service income, and the auxiliary R & D details are obtained through extraction and processing of the second project table, including:

[0028] When the block node determines that there is an upload requirement at the inquiry configuration end, an initial project table is generated and feedback to the management end, and all income data are configured for the initial project table. The income data includes multiple income information with text and / or images;

[0029] Based on the payment entity of the income information, a corresponding payment entity time axis is established, and the payment time of the same payment entity on the corresponding payment entity time axis is counted and a R & D income node is established;

[0030] The second project table is subjected to text and / or image extraction processing to obtain the corresponding income data details and set them corresponding to the corresponding R & D income nodes. The income data details at least include contracts and invoices.

[0031] Optionally, if the block server determines the data screening requirements for additional deductions uploaded by the first block node, it determines the items to be screened based on the current time point and the data axis corresponding to the first project table, and conducts data screening on the block database of the first block node and the second project table of the second block node based on the items to be screened, including:

[0032] When the block server determines the data screening requirements for additional deductions uploaded by the first block node, it determines the first screening time on the project time axis based on the current time point, and generates a screening time period based on the initial time and the first screening time of the project time axis;

[0033] Determine all sub-items corresponding to the time nodes within the screening time period as the items to be screened, and retrieve the research and development expenditures, project-related parties, texts, and / or images stored in the database for the sub-items on the project cost axis;

[0034] Determine the second block node based on the project-related party, and screen the corresponding research and development income nodes in the second project table based on the research and development expenditures, project-related parties, texts, and / or images to obtain a screening result.

[0035] Optionally, the determining the second block node based on the project-related party, and screening the corresponding research and development income nodes in the second project table based on the research and development expenditures, project-related parties, texts, and / or images to obtain a screening result includes:

[0036] Perform screening processing on the payment subject time axis based on the project-related party to obtain a target time axis;

[0037] If it is determined that the text and / or image in the first block node meet the decomposition requirements for research and development expenditures, decompose the research and development expenditures based on the text and / or image in the first block node to obtain decomposed sub-items;

[0038] Determine the research and development income nodes corresponding to the values in the target time axis based on the decomposed sub-items, and compare the text and / or image of the decomposed sub-items with the text and / or image of the research and development income nodes to obtain a screening result.

[0039] Optionally, the if it is determined that the text and / or image in the first block node meet the decomposition requirements for research and development expenditures, and decompose the research and development expenditures based on the text and / or image in the first block node to obtain decomposed sub-items includes:

[0040] Extract the contract information in the text and / or image in the first block node, and extract the payment method information and payment invoice information in the contract information;

[0041] If it is determined that the payment method information is phased payment and the images of the payment invoice information are multiple, then the amounts of the phased payment, the amounts and times of the payment invoice information are corresponded to obtain decomposed sub-items, and each decomposed sub-item has a corresponding time, amount, and invoice number.

[0042] Optionally, for the R & D income nodes corresponding to the values within the target time axis determined based on the decomposed sub-items, comparing the text and / or images of the decomposed sub-items with the text and / or images of the R & D income nodes to obtain a screening result, including:

[0043] For the R & D income nodes corresponding to the values within the target time axis determined based on the decomposed sub-items, extracting the corresponding text and / or images of the R & D income nodes to obtain the corresponding time and invoice number;

[0044] If the times correspond and the invoice numbers are the same, it is determined that the comparison screening passes and a pass result is generated; if they are inconsistent, it is determined that the comparison screening fails and a fail result is generated.

[0045] The second block node counts all the pass results and fail results of the decomposed sub-items and feeds them back to the first block node.

[0046] Optionally, the block server statistically generates a corresponding total screening result for the data screening result, and obtains and broadcasts the compliance data for additional deductions based on the total screening result, including:

[0047] The block server obtains all the pass results and fail results of each decomposed sub-item fed back by the second block nodes to the first block node and statistically counts the corresponding quantities to obtain the first statistical quantity and the second statistical quantity;

[0048] The block server obtains the number of pixel points corresponding to each decomposed sub-item fed back to obtain the first pixel point quantity of the pass result and the second pixel point quantity of the fail result;

[0049] The block server divides the project time axis into a completed project axis and an uncompleted project axis based on the current time, and comprehensively calculates the compliance data of the first block node based on the project axis completion, the first statistical quantity, the second statistical quantity, the first pixel point quantity, and the second pixel point quantity of the completed project axis and the uncompleted project axis.

[0050] Optionally, the block server divides the project time axis into a completed project axis and an uncompleted project axis based on the current time, and comprehensively calculates the compliance data of the first block node based on the project axis completion, the first statistical quantity, the second statistical quantity, the first pixel point quantity, and the second pixel point quantity of the completed project axis and the uncompleted project axis, including:

[0051] Based on the project axis completeness of the completed project axis and the uncompleted project axis, allocate corresponding compliance weights to the completed project axis and the uncompleted project axis;

[0052] Calculate the ratio of the number of second pixel points to the number of first pixel points of each project timeline to obtain the pixel point anomaly ratio. Calculate the independent anomaly factor of each project timeline based on the compliance weight and the anomaly ratio, and sum up all the independent anomaly factors to obtain the total anomaly factor;

[0053] Calculate the ratio of the second statistical quantity to the first statistical quantity to obtain the quantity anomaly ratio. After weighted processing and summing of the total anomaly factor and the quantity anomaly ratio respectively, obtain the anomaly compliance value. Subtract the anomaly compliance value from the preset value to obtain the compliance value, and determine the corresponding compliance data based on the compliance value.

[0054] Optionally, the step of allocating corresponding compliance weights to the completed project axis and the uncompleted project axis based on the project axis completeness of the completed project axis and the uncompleted project axis includes:

[0055] If it is determined that the project axis completeness is the completed project axis, add the compliance weight of the preset value;

[0056] If it is determined that the project axis completeness is the uncompleted project axis, determine the ratio of the screening time period to the total time period to obtain the time coefficient, and normalize the time coefficient to obtain the compliance weight.

[0057] Technical effects:

[0058] This solution realizes the classified storage and management of R & D behavior data and R & D service income data through block nodes. The project management module can systematically process the first project table of R & D behavior, including generating the initial project table, configuring comprehensive data, establishing the project timeline and the project cost axis, and filling the project related party details and R & D expenditure details into the corresponding data axes, realizing the orderly integration and visual presentation of R & D project data. In a complex R & D project, involving multiple sub-projects and numerous expenditure and income data, through the technical solution of the present invention, it is possible to clearly display the time nodes, cost expenditure situations of each sub-project and the relationship with the project related parties on the project timeline and the project cost axis, providing a detailed and orderly data basis for subsequent data screening and analysis, improving the efficiency and accuracy of data management, and solving the problems of scattered data and lack of effective association in the prior art.

[0059] This solution uses automation to perform text and / or image extraction processing on the first project list and the second project list, and can accurately determine key information such as the project related party, R & D expenditure, related time, and income data details of each sub - project. When extracting information from contract images and invoice images, it can accurately identify important contents such as contract numbers, contract amounts, invoice numbers, invoice amounts, etc., and store these information corresponding to the corresponding time nodes and expense nodes. For example, when processing a large number of equipment procurement contract images and invoice images, it can quickly and accurately extract relevant information, avoiding omissions and errors that may occur in manual extraction, improving the accuracy and integrity of data extraction, and effectively solving the technical problems in the prior art when dealing with mixed text and image data.

[0060] This solution constructs an intelligent data screening mechanism, determines the projects to be screened based on the current time point, project time axis, and project expense axis, and screens the corresponding R & D income nodes in the second project list, and can accurately judge the compliance of the data. During the screening process, by decomposing the R & D expenditure and making precise comparisons with the R & D income nodes, passing results and non - passing results are generated, and multi - dimensional factors such as the completion of the project axis and the abnormal ratio are comprehensively considered to calculate the compliant data. For example, when judging whether the additional deduction data of an enterprise is compliant, it can comprehensively and accurately analyze the data, avoiding misjudgments caused by single - factor judgment, providing accurate compliant data for the enterprise, ensuring that the enterprise can enjoy the additional deduction policy in compliance, improving the compliance of enterprise financial management and the accuracy of tax declaration, and solving the problems of low data screening efficiency and poor accuracy in the prior art. Brief Description of the Drawings

[0061] Figure 1 is a schematic flowchart of a data screening method for an additional deduction system provided by an embodiment of the present invention;

[0062] Figure 2 is a schematic diagram of a project expense axis provided by an embodiment of the present invention. Detailed Embodiments

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0064] See Figure 1 , which is a schematic flowchart of a data screening method for an additional deduction system provided by an embodiment of the present invention. The method includes:

[0065] S1, configure a project management module for the block node, which is used to store the first project table of the R & D behavior, and extract and process the first project table to obtain the details of project related parties and the details of R & D expenditures.

[0066] The present invention configures different modules for different block nodes, configures a project management module for the management - end block node, and configures a project inquiry module for the inquiry configuration end. The project management module is used to store the first project table of the R & D behavior, and extract and process it to obtain the details of project related parties and the details of R & D expenditures. The project inquiry module is used to store the second project table of the R & D service income, and extract and process it to obtain the details of auxiliary R & D.

[0067] In the data screening method of the additional deduction system of the present invention, step S1 is in the starting and basic construction stage of data management. By configuring a project management module for the block node and systematically processing the first project table of the R & D behavior, the key data related to the R & D project can be efficiently integrated, providing detailed and orderly data support for subsequent data screening and compliance judgment. The effective implementation of this link plays a crucial role in improving the accuracy and efficiency of the entire additional deduction management process.

[0068] In some embodiments, the configuring a project management module for the block node, which is used to store the first project table of the R & D behavior, and extract and process the first project table to obtain the details of project related parties and the details of R & D expenditures, includes:

[0069] S11, when the block node determines that the management end has an upload requirement, generate an initial project table and feedback it to the management end, and configure all expenditure data and project requirement data for the initial project table, where the expenditure data includes multiple expenditure information with text and / or images.

[0070] When the block node determines that there is an upload requirement on the management side, it will generate an initial project table and promptly feedback it to the management side. The management side will then configure comprehensive data for this initial project table, including all expenditure data and project requirement data. The expenditure data contains multiple expenditure information with text and / or images. For example, in a project involving new material research and development, the expenditure data may include the contract image of purchasing experimental equipment, in which key text information such as the supplier name, equipment model, and price is clearly presented; there is also the salary payment record of R & D personnel, presented in the form of a payslip image, which details text content such as employee name, salary amount, and payment time. These expenditure data truly reflect the investment in various aspects of the R & D project, while the project requirement data sets key indicators for the advancement of the entire project. For example, the project deadline determines the time span of the project, and the project budget clarifies the upper limit of funds. These information jointly construct the basic framework of project data management and provide an important basis for subsequent data analysis and processing.

[0071] S12, based on the project requirement data, establish a project timeline and a project cost axis corresponding to the project table. The cut-off point of the project timeline is the deadline, and the cut-off point of the project cost axis is the project budget.

[0072] Based on the project requirement data, the block node establishes a project timeline and a project cost axis corresponding to the project table. The cut-off point of the project timeline is accurately set as the deadline of the project, and the cut-off point of the project cost axis is determined as the project budget. Taking a research and development project with an expected duration of two years and a budget of 8 million yuan as an example, the cut-off point of the project timeline is the corresponding date two years after the project start date, and the cut-off point of the project cost axis is 8 million yuan. By constructing these two data axes, it is like building a two-dimensional data coordinate system for the R & D project, which can intuitively and orderly present the time dimension and cost dimension of the project. In the actual application scenario, the project timeline can clearly display the start time nodes of each sub-project in chronological order, facilitating project managers and relevant personnel to accurately grasp the project progress; the project cost axis can intuitively reflect the investment in costs for each sub-project, providing an important basis for subsequent data screening and compliance judgment.

[0073] S13, perform text and / or image extraction processing on the first project table to obtain the corresponding project related party details and R & D expenditure details, and fill them into the project timeline and the project cost axis and perform segmentation processing.

[0074] Among them, the performing text and / or image extraction processing on the first project table to obtain the corresponding project related party details and R & D expenditure details, and filling them into the project timeline and the project cost axis for segmentation processing includes:

[0075] S131. Perform text and / or image extraction processing on the first project list to determine the project related parties, R & D expenditures, and related times of each sub - project. Each sub - project has corresponding text and / or images.

[0076] The system uses the text recognition OCR technology and image analysis algorithms in the existing technology to extract data from the first project list. In a large R & D project involving multiple sub - projects, for the equipment procurement sub - project, the system can accurately extract from the procurement contract image and related text descriptions that the project related party is a well - known equipment supplier, the R & D expenditure is 1.5 million yuan, and the related time is March 2025. Similarly, for the personnel training sub - project, it can be determined that the project related party is a professional training company, the R & D expenditure is 300,000 yuan, and the related time is May 2024, etc. In this way, the system comprehensively and accurately determines the key information of each sub - project, providing a detailed data source for subsequent data sorting and analysis.

[0077] S132. Based on the related time, establish time nodes corresponding to each sub - project on the project timeline, and store the corresponding text and / or images, project related parties in the first project list in correspondence with the respective time nodes.

[0078] According to the related time of each sub - project extracted, the system accurately establishes corresponding time nodes on the project timeline. For example, for the above - mentioned equipment procurement sub - project, a time node is established at the position corresponding to March 2024 on the project timeline, and the image of the equipment procurement contract, the detailed information of the equipment supplier, etc. are stored in correspondence with this time node. In this way, through the project timeline, it is possible to clearly view the sub - project - related information occurring at different time points, providing an intuitive perspective for project progress tracking and analysis.

[0079] S133. Based on the order of the project related parties on the project timeline, establish stepped cost nodes corresponding to each sub - project for the R & D expenditures of each project related party on the project cost axis, and store the corresponding text and / or images, project related parties in the first project list in correspondence with the respective stepped cost nodes.

[0080] This step plays a key role in accurately associating and visually presenting the R & D expenditure data with the project cost axis in the data processing flow of the entire project management module. By establishing stepped cost nodes on the project cost axis and storing the corresponding text and / or images and project related party information, it is possible to intuitively display the cost expenditure of each sub - project and its proportion in the total budget, providing a clear and accurate data basis for subsequent data screening and project cost analysis.

[0081] In some embodiments, for the order of project related parties based on the project timeline, stepwise cost nodes corresponding to each sub - project are established on the project cost axis for the R & D expenditures of each project related party, and the corresponding text and / or images in the first project table, and the project related parties are stored corresponding to the respective stepwise cost nodes, including:

[0082] S1331. Calculate the unit amount corresponding to each pixel point based on the pixel point length of the project cost axis and the project budget at the cut - off point.

[0083] During the construction of the project cost axis, the system will perform accurate calculations based on the pixel point length of the project cost axis and the project budget at the cut - off point. The project cost axis is essentially a virtual coordinate axis that intuitively shows the cost distribution in units of pixel points. For example, in a project, the pixel point length of the project cost axis is set to 1500 pixel points, and the project budget is 7.5 million yuan. Then, through a simple division operation, that is, 7500000÷1500 = 5000 yuan, the system can obtain that the unit amount corresponding to each pixel point is 5000 yuan. The determination of this unit amount provides a key measurement standard for accurately mapping the R & D expenditures to the project cost axis subsequently.

[0084] S1332. Extract each project related party and the corresponding R & D expenditure in sequence according to the order of project related parties on the project timeline, and calculate the number of pixel points corresponding to the project related party based on the R & D expenditure and the unit amount.

[0085] The system will extract and analyze each project related party and its corresponding R & D expenditure in sequence according to the order of project related parties on the project timeline. Taking a R & D project with multiple sub - projects as an example, on the project timeline, the first project related party that appears is a certain software supplier, and its corresponding R & D expenditure is 1 million yuan. According to the previously calculated unit amount of 5000 yuan, through the division operation 1000000÷5000 = 200, the system can accurately calculate that the number of pixel points corresponding to this software supplier is 200. In this way, the system converts the R & D expenditures of each project related party into quantifiable pixel point numbers on the project cost axis, providing a direct data basis for determining the stepwise cost nodes subsequently.

[0086] S1333. Determine the stepwise cost nodes corresponding to each sub - project in sequence based on the number of pixel points, and establish a storage space corresponding to each stepwise cost node to store the corresponding text and / or images and project related parties.

[0087] Among them, determining the stepped cost nodes corresponding to each sub-item in sequence based on the number of pixel points, and establishing storage spaces corresponding to each stepped cost node to store the corresponding text and / or images and project related parties includes:

[0088] If it is determined that there are no stepped cost nodes corresponding to other sub-items in the front part of the project cost axis, then starting from the first point of the project cost axis, the stepped cost nodes for this time are sequentially counted and selected according to the number of pixel points.

[0089] When determining the stepped cost nodes, the system will make an intelligent judgment based on the situation in the front part of the project cost axis. If there are no stepped cost nodes corresponding to other sub-items in the front part of the project cost axis, this means that the current sub-item is the first item for which cost marking is carried out on the project cost axis. At this time, the system will start from the first point of the project cost axis and sequentially count according to the number of pixel points calculated previously. For example, for the 200 pixel points corresponding to the above software vendor, the system starts from the starting point of the project cost axis and sequentially selects 200 pixel points, and the range of these selected pixel points is determined as the stepped cost nodes for this time.

[0090] If it is determined that there are stepped cost nodes corresponding to other sub-items in the front part of the project cost axis, then starting from the last point of the last other sub-item in the project cost axis, the stepped cost nodes for this time are sequentially counted and selected according to the number of pixel points.

[0091] If there are already stepped cost nodes corresponding to other sub-items in the front part of the project cost axis, the system will start from the last point of the last other sub-item in the current project cost axis. Suppose there is already a stepped cost node for a sub-item of equipment procurement in the front, and the pixel point position of its last point is 300, and the number of pixel points corresponding to the current software vendor is 200, then the system will start from pixel point position 301 and sequentially count 200 pixel points to determine the stepped cost nodes for the software vendor sub-item for this time.

[0092] Calculate the proportion of the number of pixel points of each sub-item in the total number of pixel points and store it in the corresponding storage space.

[0093] After determining the tiered cost node for each sub-project, the system will establish a storage space corresponding to each tiered cost node. For software supplier sub-projects, the system will store the relevant information of the software supplier, such as the text and image of the software purchase contract (including detailed information such as software name, version, purchase amount, etc.), as well as the name and contact information of the project related party (software supplier), in the storage space corresponding to the tiered cost node. At the same time, the system will also calculate the proportion of the number of pixels of each sub-project to the total number of pixels and store it in the corresponding storage space. Continuing with the above example, the total number of pixels is 1500, and the number of pixels of the software supplier sub-project is 200. Then the proportion of the number of pixels of the sub-project to the total number of pixels is 200÷1500≈13.33%. The system will store this proportion information in the storage space corresponding to the tiered cost node of the sub-project. Through this storage method, not only can the position and cost proportion of each sub-project on the project cost axis be intuitively displayed, but also the subsequent query and analysis of the cost data of each sub-project can be facilitated.

[0094] It is worth mentioning that see Figure 2 The above-mentioned project cost axis can be in the shape of an ascending ladder, and a ladder cost node is the dividing point of a ladder step.

[0095] S2, configures a project query module for the block node, which is used to store the second project table of R&D service income, and extracts and processes the second project table to obtain auxiliary R&D details.

[0096] In the additional deduction system of the present invention, step S2 realizes the effective management and analysis of R&D service income data by configuring the project query module for the block node, which is essential for comprehensive and accurate screening of additional deduction data. The second project table data stored and processed by this module is interrelated and mutually verified with the R&D behavior data in the project management module, providing income-end data support for the subsequent accurate determination of compliant data that complies with the additional deduction policy.

[0097] In some embodiments, the project query module is configured for the block node to store the second project table of R&D service income, and the second project table is extracted and processed to obtain the auxiliary R&D details, including:

[0098] S21, when the block node determines that the query configuration end has an upload requirement, it generates an initial project list and feeds it back to the management end, and configures all income data for the initial project list, wherein the income data includes a plurality of income information with text and / or images.

[0099] The block node in this step corresponds to the inquiry configuration end. When the block node detects an upload requirement generated by the inquiry configuration end, it will start the generation process of the initial project table. Taking an enterprise engaged in software R & D and providing related technical services as an example, when the financial department or business department of the enterprise issues a data upload instruction at the inquiry configuration end, the block node generates the initial project table and feeds it back to the management end. The management end then configures all the revenue data related to the R & D service revenue for this table. These revenue data are in various forms. For example, the electronic document image of a software R & D service contract, which contains text information such as customer name, service content, contract amount, payment method, etc.; there are also scanned images of invoices, as well as corresponding text information such as invoice number, invoicing date, amount, etc. By comprehensively configuring these revenue data, it ensures that the second project table can completely carry the R & D service revenue information of the corresponding project of the enterprise, laying a solid data foundation for subsequent data processing and analysis.

[0100] S22. Based on the payment entity in the revenue information, establish the corresponding payment entity timeline, count the payment times of the same payment entity on the corresponding payment entity timeline, and establish R & D revenue nodes.

[0101] The system constructs the corresponding payment entity timeline based on the payment entity in the revenue information. Continuing with the above software R & D enterprise as an example, assuming that the enterprise provides R & D services for multiple customers, and customer A has made multiple payments at different times. The system will establish an independent payment entity timeline with customer A as the unit. On this timeline, the system will carefully count the times when customer A pays the R & D fees each time, such as March 15, 2025, May 20, 2025, etc., and establish R & D revenue nodes at the corresponding time points. In this way, the payment time information of each payment entity is sorted and presented intuitively, facilitating subsequent tracking and analysis of the revenue situations of different payment entities.

[0102] S23. Perform text and / or image extraction processing on the second project table to obtain the corresponding revenue data details and set them corresponding to the corresponding R & D revenue nodes. The revenue data details at least include contracts and invoices.

[0103] The system uses the text recognition OCR technology and image analysis algorithm in the existing technology to deeply extract data from the second project list. During the extraction process, the system can accurately identify key information such as contract numbers, contract amounts, invoice numbers, invoice amounts, etc. from contract images and invoice images, and form corresponding income data details. For example, from the contract image of the software R & D service contract signed with customer A, the contract number "SW2024001" and the contract amount of 500,000 yuan are extracted; from the corresponding invoice image, the invoice number "00123456" and the invoice amount of 200,000 yuan (assuming it is an invoice for part of the payment of this contract) are extracted. Then, the system precisely sets the corresponding relationship between these income data details and the corresponding R & D income nodes established on the payment subject time axis. For example, the contract and invoice information related to customer A mentioned above is associated and stored with the R & D income node corresponding to March 15, 2024 on the payment subject time axis of customer A. Through this corresponding setting, the income data is closely combined with the time nodes, providing a clear and accurate data association relationship for subsequent data screening and analysis, and facilitating quick query and comparison of the income details of each payment subject at different time points.

[0104] S3. If the block server determines the data screening requirement for additional deductions uploaded by the first block node, it determines the items to be screened based on the current time point and the data axis corresponding to the first project list, and conducts data screening based on the block database of the first block node and the second project list of the second block node determined by the items to be screened.

[0105] This step is a core operation link in the data screening method of the entire additional deduction system. It is closely related to the data stored in the project management module and the project inquiry module. By accurately analyzing the time axis and comparing data, it determines the data that meets the additional deduction conditions, provides a direct basis for generating compliant data, and plays a decisive role in achieving the accuracy and efficiency of additional deduction management.

[0106] In some embodiments, if the block server determines the data screening requirement for additional deductions uploaded by the first block node, it determines the items to be screened based on the current time point and the data axis corresponding to the first project list, and conducts data screening based on the block database of the first block node and the second project list of the second block node determined by the items to be screened, including:

[0107] S31. When the block server determines the data screening requirement for additional deductions uploaded by the first block node, it determines the first screening time on the project time axis based on the current time point, and generates a screening time period based on the initial time and the first screening time of the project time axis.

[0108] When the block server detects the data screening demand for additional deductions uploaded by the first block node, the timeline analysis process will be started. For example, on November 15, 2024, the first block node of a certain enterprise initiates a data screening request, and the block server obtains the current time point, which is November 15, 2024. Based on the previously constructed project timeline, assuming that the initial time of the project timeline is January 1, 2024, the block server determines November 15, 2024 as the first screening time. Then, through a simple time range definition, the screening time period is generated from January 1, 2024 to November 15, 2024. The determination of this screening time period provides a clear time range for the subsequent screening of related sub-projects, ensuring that only the R&D project data within this time period is screened, which improves the pertinence and efficiency of data processing.

[0109] S32, determining the sub-projects corresponding to all time nodes within the screening time period as the projects to be screened, and retrieving the R&D expenditures, project related parties, texts and / or images of the sub-projects respectively stored in the database in the project cost axis.

[0110] Based on the generated screening time period, the system will accurately locate the sub-projects corresponding to all time nodes within the time period on the project timeline. Continuing with the above project as an example, during the period from January 1, 2024 to November 15, 2024, there are multiple sub-projects on the project timeline, such as the equipment procurement sub-project (associated time is March 2024) and the personnel training sub-project (associated time is May 2024). These sub-projects are the projects to be screened. Subsequently, the system will retrieve key information such as R&D expenditures, project related parties, text and / or images stored in the project cost axis of these sub-projects to be screened from the database.

[0111] S33, determining the second block node based on the project related parties, and screening the corresponding R&D income nodes in the second project table based on the R&D expenditure, project related parties, text and / or image to obtain a screening result.

[0112] The determining of the second block node based on the project related party and the screening of the corresponding R&D income node in the second project table based on the R&D expenditure, project related party, text and / or image to obtain the screening result include:

[0113] S331, based on the project related party, the payment subject timeline is screened and processed to obtain a target timeline.

[0114] Based on the information of project-related parties, the system filters in the payment entity timeline constructed in the project inquiry module. Suppose in the project to be filtered, the project-related party of the equipment procurement sub-project is a certain equipment supplier, and this equipment supplier exists as a payment entity in the project inquiry module. The system will filter out the payment entity timeline corresponding to this equipment supplier from all payment entity timelines and determine it as the target timeline. All R & D service income payment times and related information of this equipment supplier as a payment entity are recorded on this target timeline, providing an accurate source of income data for subsequent data comparison.

[0115] S332. If it is determined that the text and / or image within the first block node meet the requirements for decomposing R & D expenditures, then decompose the R & D expenditures based on the text and / or image within the first block node to obtain decomposition sub-items.

[0116] In some embodiments, the step of if it is determined that the text and / or image within the first block node meet the requirements for decomposing R & D expenditures, then decompose the R & D expenditures based on the text and / or image within the first block node to obtain decomposition sub-items includes:

[0117] S3321. Extract the contract information from the text and / or image within the first block node, and extract the payment method information and payment invoice information from the contract information.

[0118] The system extracts the text and / or image related to the project to be filtered within the first block node. Taking the equipment procurement sub-project as an example, contract information such as contract number, procurement amount, payment method, etc., and payment invoice information such as invoice number, invoiced amount, invoicing time, etc. are extracted from the text and images of its procurement contract. For example, the contract information shows that the payment method is phased payment, and the payment invoice information includes multiple invoice images, with each invoice corresponding to a different payment amount and time.

[0119] S3322. If it is determined that the payment method information is phased payment and the images of the payment invoice information are multiple, then correspond the amounts of the phased payments, the amounts and times of the payment invoice information to obtain decomposition sub-items, and each decomposition sub-item has a corresponding time, amount, and invoice number.

[0120] When the system determines that the payment method information is installment payment and the images of the payment invoice information are multiple, it will accurately match the amounts of the installment payment, the amounts and times of the payment invoice information. Suppose the total amount of the equipment procurement contract is 1.5 million yuan, which is paid in three installments. The first payment of 500,000 yuan is made in March 2024, corresponding to the invoice number 00123456; the second payment of 500,000 yuan is made in June 2024, corresponding to the invoice number 00234567; the third payment of 500,000 yuan is made in September 2024, corresponding to the invoice number 00345678. The system will organize this information to generate three decomposed sub-items, and each decomposed sub-item contains the corresponding time (March 2024, June 2024, September 2024), amount (500,000 yuan, 500,000 yuan, 500,000 yuan), and invoice number (00123456, 00234567, 00345678). In this way, the complex R & D expenditure information is refined and decomposed, providing detailed data units for the subsequent precise comparison with the R & D income nodes.

[0121] S333. Determine the R & D income nodes corresponding to the values within the target time axis based on the decomposed sub-items, and compare the text and / or image of the decomposed sub-items with the text and / or image of the R & D income nodes to obtain the screening result.

[0122] In some embodiments, the determining the R & D income nodes corresponding to the values within the target time axis based on the decomposed sub-items and comparing the text and / or image of the decomposed sub-items with the text and / or image of the R & D income nodes to obtain the screening result includes:

[0123] S3331. Determine the R & D income nodes corresponding to the values within the target time axis based on the decomposed sub-items, and extract the text and / or image corresponding to the R & D income nodes to obtain the corresponding time and invoice number.

[0124] Based on the generated decomposed sub-items, the system searches for the R & D income nodes corresponding to the values within the target time axis. For example, for the first decomposed sub-item (March 2024, 500,000 yuan, 00123456), the system searches on the time axis of the payment entity of the equipment supplier for the R & D income node around March 2024 with an amount of 500,000 yuan and a matching invoice number. When the corresponding R & D income node is found, extract the text and / or image corresponding to the node to obtain information such as the corresponding time and invoice number.

[0125] S3332. If the times correspond and the invoice numbers are the same, it is determined that the comparison and screening pass and a pass result is generated; if they are inconsistent, it is determined that the comparison and screening fails and a fail result is generated.

[0126] The system strictly compares the information such as the time of the decomposed sub-items, invoice numbers, etc. with the information extracted from the R & D income nodes. If the time of the decomposed sub-items is consistent with the time of the R & D income nodes and the invoice numbers also match exactly, it is judged that the comparison and screening pass, and a pass result is generated; if the time is inconsistent or the invoice numbers are inconsistent, it is judged that the comparison and screening fails, and a fail result is generated. For example, if the time and invoice number of the first decomposed sub-item exactly match the information of the R & D income node, the system generates a pass result; if the invoice numbers are inconsistent, the system generates a fail result.

[0127] S3333, the second block node counts all the pass results and fail results of the decomposed sub-items and feeds them back to the first block node.

[0128] The second block node is responsible for counting all the pass results and fail results of the decomposed sub-items and feeding these results back to the first block node. Suppose three decomposed sub-items are generated for the equipment procurement sub-project. After comparison, two of the decomposed sub-items pass the screening and one fails the screening. The second block node will feed this result back to the first block node. Through this result feedback mechanism, the first block node can comprehensively understand the screening situation of each project to be screened, providing accurate data support for generating compliance data subsequently.

[0129] S4, the block server statistically generates the corresponding total screening result based on the results of the data screening, obtains the compliance data for additional deductions based on the total screening result and broadcasts it.

[0130] In the data screening process of the entire additional deduction system, step S4 is a key link in the final data summary and compliance determination. It analyzes the screening results fed back by each second block node, comprehensively considers the completion of the project time axis and the abnormal ratios of different types, accurately calculates the compliance data of the first block node, and broadcasts it, providing a clear basis for additional deduction data for enterprises, directly affecting the tax treatment and financial management decisions of enterprises.

[0131] In some embodiments, the block server statistically generates the corresponding total screening result based on the results of the data screening, obtains the compliance data for additional deductions based on the total screening result and broadcasts it, including:

[0132] S41, the block server obtains all the pass results and fail results of each decomposed sub-item fed back by the second block nodes to the first block node and statistically counts the corresponding quantities to obtain the first statistical quantity and the second statistical quantity.

[0133] After all the second block nodes have completed the screening of each decomposed item of the first block node and fed back the passing results and non-passing results, the block server starts the statistical process. For example, in a scenario involving the screening of multiple R & D project sub-items, the second block node A feeds back the passing result of a certain equipment procurement decomposed item, and the second block node B feeds back the non-passing result of a certain technical consulting service decomposed item, etc. The block server will classify and count these feedback results, count the number of all passing results as the first statistical quantity, and count the number of all non-passing results as the second statistical quantity. Suppose in a screening, the number of passing results received is 15, and the number of non-passing results is 5, then the first statistical quantity is 15, and the second statistical quantity is 5. These quantity statistical results are important basic data for subsequent analysis of data screening quality and compliance.

[0134] S42. The block server obtains the number of pixel points corresponding to each decomposed item fed back, and obtains the first pixel point quantity of the passing results and the second pixel point quantity of the non-passing results.

[0135] The block server further obtains the pixel point quantity information corresponding to each decomposed item. Since when constructing the project cost axis in the early stage, the R & D expenditure of each decomposed item corresponds to a certain number of pixel points, by counting these pixel point quantities, the data screening situation can be further analyzed from the perspective of cost distribution. For example, for the decomposed item that passes the screening, the total number of corresponding pixel points is 1200 (assumed), and this quantity is the first pixel point quantity of the passing results; for the decomposed item that fails the screening, the total number of corresponding pixel points is 300 (assumed), and this quantity is the second pixel point quantity of the non-passing results.

[0136] S43. The block server divides the project time axis into a completed project axis and an uncompleted project axis based on the current time, and comprehensively calculates the compliance data of the first block node based on the project axis completeness, the first statistical quantity, the second statistical quantity, the first pixel point quantity, and the second pixel point quantity of the completed project axis and the uncompleted project axis.

[0137] This solution will summarize and calculate multi-dimensional data to obtain the compliance data of the first block node.

[0138] Among them, the block server divides the project time axis into a completed project axis and an uncompleted project axis based on the current time, and comprehensively calculates the compliance data of the first block node based on the project axis completeness, the first statistical quantity, the second statistical quantity, the first pixel point quantity, and the second pixel point quantity of the completed project axis and the uncompleted project axis, including:

[0139] S431. Assign corresponding compliance weights to the completed project axis and the incomplete project axis based on the completion of the project axis of the completed projects and the incomplete projects.

[0140] This step is mainly used to assign compliance weights.

[0141] In some embodiments, the assigning corresponding compliance weights to the completed project axis and the incomplete project axis based on the completion of the project axis of the completed projects and the incomplete projects includes:

[0142] S4311. If it is determined that the project axis completion is the completed project axis, add a compliance weight with a preset value.

[0143] When the block server determines that the project axis completion is the completed project axis, it will add a compliance weight with a preset value. The preset value is usually determined comprehensively based on various factors such as the business characteristics of the enterprise, industry standards, and the requirements of the additional deduction policy. For example, in a certain industry, based on past experience and policy guidance, the compliance weight of the completed project axis is preset to 0.8. This means that the completed projects have a relatively high credibility and weight in the compliance determination because the data of the completed projects is relatively more complete and certain.

[0144] S4312. If it is determined that the project axis completion is the incomplete project axis, determine the proportion of the screening time period to the total time period to obtain a time coefficient, and normalize the time coefficient to obtain a compliance weight.

[0145] If it is determined that the project axis completion is the incomplete project axis, the system will first determine the proportion of the screening time period to the total time period to obtain a time coefficient. Suppose the total planned duration of a project is 12 months, and the current screening time period is from the 1st month to the 8th month after the start of the project, then the time coefficient is 8÷12≈0.67. Then, normalize this time coefficient, that is, adjust it to a reasonable weight range. For example, after normalizing the time coefficient, the obtained compliance weight is 0.6. In this way, a relatively reasonable compliance weight is assigned to the incomplete project axis according to the actual progress of the project.

[0146] S432. Calculate the ratio of the number of second pixel points to the number of first pixel points of each project time axis to obtain a pixel point anomaly ratio, calculate the independent anomaly factor of each project time axis based on the compliance weight and the anomaly ratio, and sum up all the independent anomaly factors to obtain a total anomaly factor.

[0147] The system calculates the ratio of the number of second pixels to the number of first pixels for each project timeline to obtain the pixel anomaly ratio. For example, for a completed project timeline, assuming the number of first pixels is 800 and the number of second pixels is 200, then the pixel anomaly ratio is 200÷800 = 0.25; for an unfinished project timeline, assuming the number of first pixels is 400 and the number of second pixels is 100, then the pixel anomaly ratio is 100÷400 = 0.25. Then, based on the previously assigned compliance weights and the calculated pixel anomaly ratio, the independent anomaly factor for each project timeline is calculated. For example, for a completed project timeline, the independent anomaly factor is 0.8×0.25 = 0.2; for an unfinished project timeline, the independent anomaly factor is 0.6×0.25 = 0.15. Finally, the independent anomaly factors of all project timelines are summed to obtain the total anomaly factor, that is, 0.2 + 0.15 = 0.35. This total anomaly factor reflects the degree of anomaly in the pixel distribution level of the overall data screening result.

[0148] S433, calculate the ratio of the second statistical quantity to the first statistical quantity to obtain the quantity anomaly ratio, perform weighted processing on the total anomaly factor and the quantity anomaly ratio respectively and sum them to obtain the anomaly compliance value, subtract the anomaly compliance value from the preset value to obtain the compliance value, and determine the corresponding compliance data based on the compliance value.

[0149] Calculate the ratio of the second statistical quantity to the first statistical quantity to obtain the quantity anomaly ratio. In the previous example, the first statistical quantity is 15 and the second statistical quantity is 5, then the quantity anomaly ratio is 5÷15≈0.33. Then, perform weighted processing on the total anomaly factor and the quantity anomaly ratio respectively. Assuming that based on experience and data analysis, the weight of the total anomaly factor is set to 0.6 and the weight of the quantity anomaly ratio is set to 0.4. Then the anomaly compliance value is 0.35×0.6 + 0.33×0.4 =0.342. Assuming the preset value is 1, then the compliance value is 1 - 0.342 = 0.658. Finally, determine the corresponding compliance data based on this compliance value, such as the amount eligible for additional deductions. For example, if the total R & D expenditure of an enterprise is 10 million yuan, the amount eligible for additional deductions calculated according to the compliance value is 1000×0.658 = 6.58 million yuan. The block server broadcasts these compliance data, and the relevant departments of the enterprise can make accurate tax declarations and financial management decisions based on these data.

[0150] The present invention also provides a storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it is used to implement the methods provided by the above various embodiments.

[0151] Among them, the storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, the storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also exist as discrete components in a communication device. The storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0152] The present invention also provides a program product, which includes execution instructions stored in a storage medium. At least one processor of the device can read the execution instructions from the storage medium, and the execution of the execution instructions by at least one processor enables the device to implement the methods provided by the above various embodiments.

[0153] In the above embodiments of the terminal or the server, it should be understood that the processor can be a central processing unit (CPU for short), and can also be other general-purpose processors, digital signal processors (DSP for short), application specific integrated circuits (ASIC for short), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the present invention can be directly embodied as being completed by the execution of a hardware processor, or by a combination of hardware and software modules in the processor.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A data screening method for a additional deduction system, characterized in that Including: Configure a project management module for the block node, which is used to store the first project table of R & D behaviors, and extract and process the first project table to obtain the details of project related parties and R & D expenditure details; Configure a project inquiry module for the block node, which is used to store the second project table of R & D service income, and extract and process the second project table to obtain the details of auxiliary R & D; If the block server determines the data screening requirements for additional deductions uploaded by the first block node, it determines the projects to be screened based on the current time point and the data axis corresponding to the first project table, and determines the block database of the first block node and the second project table of the second block node for data screening based on the projects to be screened, including: Determine all sub-projects corresponding to all time nodes within the screening time period as the projects to be screened, and retrieve the R & D expenditure, project related parties, text, and / or images stored in the database at the project cost axis of the sub-projects; Determine the second block node based on the project related parties, and screen the corresponding R & D income nodes in the second project table based on the R & D expenditure, project related parties, text, and / or images to obtain the screening results; The block server statistically generates the corresponding total screening results based on the data screening results, and obtains the compliant data for additional deductions based on the total screening results and broadcasts them, including: The block server obtains the passing results and non-passing results of each decomposed sub-item fed back by all second block nodes to the first block node and statistically obtains the corresponding first statistical quantity and second statistical quantity; The block server obtains the number of pixel points corresponding to each decomposed sub-item fed back, and obtains the first pixel point quantity of the passing result and the second pixel point quantity of the non-passing result; The block server divides the project time axis into a completed project axis and an uncompleted project axis based on the current time, and comprehensively calculates the compliant data of the first block node based on the project axis completeness, first statistical quantity, second statistical quantity, first pixel point quantity, and second pixel point quantity of the completed project axis and the uncompleted project axis, including: Based on the project axis completeness of the completed project axis and the uncompleted project axis, allocate corresponding compliance weights to the completed project axis and the uncompleted project axis; Calculate the ratio of the second pixel point quantity to the first pixel point quantity of each project time axis to obtain the pixel point anomaly ratio, calculate the independent anomaly factor of each project time axis based on the compliance weight and the anomaly ratio, and sum all the independent anomaly factors to obtain the total anomaly factor; Calculate the ratio of the second statistical quantity to the first statistical quantity to obtain the quantity anomaly ratio, perform weighted processing on the total anomaly factor and the quantity anomaly ratio respectively and sum them to obtain the anomaly compliance value, subtract the anomaly compliance value from the preset value to obtain the compliance value, and determine the corresponding compliance data based on the compliance value.

2. The data screening method for the additional deduction system according to claim 1, wherein The configuration of the project management module for the block node, which is used to store the first project table of R & D behaviors, and extract and process the first project table to obtain the details of project related parties and R & D expenditure details, includes: When the block node determines that the management end has an upload requirement, it generates an initial project table and feeds it back to the management end, and configures all expenditure data and project requirement data for the initial project table. The expenditure data includes multiple expenditure information with text and / or images; Based on the project requirement data, a project timeline and a project cost axis corresponding to the project table are established. The cut-off point of the project timeline is the cut-off time, and the cut-off point of the project cost axis is the project budget; Perform text and / or image extraction processing on the first project table to obtain the corresponding project related party details and R & D expenditure details, and fill them into the project timeline and the project cost axis and perform segmentation processing.

3. The data screening method for the additional deduction system according to claim 2, wherein, The performing text and / or image extraction processing on the first project table to obtain the corresponding project related party details and R & D expenditure details, and filling them into the project timeline and the project cost axis for segmentation processing includes: Perform text and / or image extraction processing on the first project table to determine the project related parties, R & D expenditures and related times of each sub-project. Each sub-project has corresponding text and / or images; Based on the related time, establish time nodes corresponding to each sub-project on the project timeline, and store the corresponding text and / or images and project related parties in the first project table corresponding to the time nodes; Based on the order of the project related parties on the project timeline, establish stepped cost nodes corresponding to each sub-project for the R & D expenditures of each project related party on the project cost axis, and store the corresponding text and / or images and project related parties in the first project table corresponding to the stepped cost nodes.

4. The data screening method for the additional deduction system according to claim 3, wherein, The establishing stepped cost nodes corresponding to each sub-project for the R & D expenditures of each project related party on the project cost axis based on the order of the project related parties on the project timeline, and storing the corresponding text and / or images and project related parties in the first project table corresponding to the stepped cost nodes includes: Calculate the unit amount corresponding to each pixel point based on the pixel point length of the project cost axis and the project budget at the cut-off point; Extract each project related party and the corresponding R & D expenditure in sequence according to the order of the project related parties on the project timeline, and calculate the number of pixel points corresponding to the project related party based on the R & D expenditure and the unit amount; Based on the number of pixel points, sequentially determine the stepped cost nodes corresponding to each sub-project, and establish a storage space corresponding to each stepped cost node to store the corresponding text and / or images and project related parties.

5. The data screening method for the additional deduction system according to claim 4, wherein, The sequentially determining the stepped cost nodes corresponding to each sub-project based on the number of pixel points, and establishing a storage space corresponding to each stepped cost node to store the corresponding text and / or images and project related parties includes: If it is determined that there are no stepped cost nodes corresponding to other sub - projects in the front part of the project cost axis, then starting from the first point of the project cost axis, count and select in sequence according to the number of pixel points to obtain the stepped cost nodes for this time; If it is determined that there are stepped cost nodes corresponding to other sub - projects in the front part of the project cost axis, then starting from the last point of the last other sub - project in the project cost axis, count and select in sequence according to the number of pixel points to obtain the stepped cost nodes for this time; Calculate the proportion of the number of pixel points of each sub - project to the total number of pixel points and store it in the corresponding storage space.

6. The data screening method for the additional deduction system according to claim 1, wherein A project inquiry module is configured for the block node, which is used to store the second project form of the R & D service income, and extract and process the second project form to obtain the auxiliary R & D details, including: When the block node determines that there is an upload requirement at the inquiry configuration end, it generates an initial project form and feeds it back to the management end, and configures all income data for the initial project form. The income data includes multiple income information with text and / or images; Based on the payment entity of the income information, establish a corresponding payment entity time axis, count the payment time of the same payment entity on the corresponding payment entity time axis, and establish R & D income nodes; Perform text and / or image extraction processing on the second project form to obtain the corresponding income data details and set them corresponding to the corresponding R & D income nodes. The income data details include at least contracts and invoices.

7. The data screening method for the additional deduction system according to claim 6, wherein If the block server determines the data screening requirement for additional deduction uploaded by the first block node, then based on the current time point and the data axis corresponding to the first project form, determine the project to be screened, and perform data screening based on the block database of the first block node and the second project form of the second block node for the project to be screened, including: When the block server determines the data screening requirement for additional deduction uploaded by the first block node, then based on the current time point, determine the first screening time on the project time axis, and generate a screening time period based on the initial time and the first screening time of the project time axis.

8. The data screening method for the additional deduction system according to claim 7, wherein Determine the second block node based on the project related party, and screen the corresponding R & D income nodes in the second project form based on the R & D expenditure, project related party, text and / or images to obtain a screening result, including: Perform screening processing on the payment entity time axis based on the project related party to obtain the target time axis; If it is determined that the text and / or images in the first block node meet the decomposition requirements for R & D expenditure, then decompose the R & D expenditure based on the text and / or images in the first block node to obtain decomposition sub - items; Based on the decomposition sub - items, determine the R & D income nodes with corresponding values in the target time axis, and compare the text and / or images of the decomposition sub - items with the text and / or images of the R & D income nodes to obtain the screening result.

9. The data screening method for the additional deduction system according to claim 8, wherein If it is determined that the text and / or image in the first block node meet the requirements for the decomposition of R & D expenditures, then based on the text and / or image in the first block node, the decomposed sub-items of R & D expenditures are obtained, including: Extract the contract information in the text and / or image in the first block node, and extract the payment method information and payment invoice information in the contract information; If it is determined that the payment method information is phased payment and the image of the payment invoice information is multiple, then the amounts of phased payment, the amounts and times of the payment invoice information are corresponded to obtain the decomposed sub-items, and each decomposed sub-item has the corresponding time, amount and invoice number.

10. The data screening method for the additional deduction system according to claim 9, wherein Based on the decomposed sub-items, the R & D income nodes corresponding to the values in the target time axis are determined, and the text and / or image of the decomposed sub-items are compared with the text and / or image of the R & D income nodes to obtain the screening result, including: Based on the decomposed sub-items, the R & D income nodes corresponding to the values in the target time axis are determined, and the text and / or image corresponding to the R & D income nodes are extracted to obtain the corresponding time and invoice number; If the time corresponds and the invoice numbers are the same, it is determined that the comparison screening passes and a passing result is generated. If they are not the same, it is determined that the comparison screening fails and a non-passing result is generated; The second block node counts all the passing results and non-passing results of the decomposed sub-items and feeds them back to the first block node.

11. The data screening method for the additional deduction system according to claim 1, wherein Based on the project axis completeness of the completed project axis and the uncompleted project axis, the corresponding compliance weights are assigned to the completed project axis and the uncompleted project axis, including: If it is determined that the project axis completeness is the completed project axis, then a compliance weight of a preset value is added; If it is determined that the project axis completeness is the uncompleted project axis, then the proportion of the screening time period in the total time period is determined to obtain the time coefficient, and the compliance weight is obtained after normalizing the time coefficient.

Images