Financial bill and bill full-automatic matching method and device, equipment and medium

Abstract

The invention discloses a full-automatic matching method, device, equipment and medium for financial bills and receipts under a fuzzy condition, and the method comprises the steps: obtaining financial bill information and receipt information, and carrying out the feature extraction of the financial bill information and the receipt information, obtaining a bill characteristic value corresponding to each bill and a bill characteristic value corresponding to each bill; according to each document feature value, generating a document abstract corresponding to each document; according to the bill feature values, the bill feature values and the bill abstracts, matching the bills with the bills one by one, and outputting a first matching result; and if the unmatched bills and the unmatched bills exist, performing fuzzy matching according to the corresponding bill feature values and the corresponding bill feature values, and outputting a second matching result. According to the method, a two-step matching method is introduced, so that the accuracy of full-automatic matching of financial bills and receipts under fuzzy conditions is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of financial management, and in particular to a method, device, equipment and medium for fully automatic matching of financial bills and documents. Background Art

[0002] Matching financial invoices with documents is a necessary and time-consuming process in accounting practice. Currently, this process is largely manual, requiring the individual bank receipt numbers to be found, followed by printing and sorting. These steps are tedious and time-consuming. Consequently, automated matching systems have been developed both domestically and internationally. These systems automatically identify and match receipts with documents, reducing manual intervention and significantly improving efficiency.

[0003] However, when the information on bills and documents is unclear, existing automatic bill-to-document matching systems require technical modifications such as QR codes to achieve automatic matching. This requires significant modification costs and changes to existing operating practices. Therefore, achieving more efficient and convenient automatic bill-to-document matching under ambiguous conditions has become a technical bottleneck that urgently needs to be addressed in the field of financial management technology. Summary of the Invention

[0004] The present invention provides a method, device, equipment and medium for fully automatic matching of financial bills and documents under fuzzy conditions, which can achieve high accuracy of fully automatic matching of financial bills and documents under fuzzy conditions.

[0005] In a first aspect, an embodiment of the present invention provides a method for fully automatic matching of financial bills and documents under fuzzy conditions, comprising:

[0006] Obtain financial bill information and document information, and perform feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document;

[0007] Generating a document summary corresponding to each document according to each document characteristic value;

[0008] Matching each of the bills with each of the documents according to the characteristic values ​​of each bill, the characteristic values ​​of each document, and the summary of each document, and outputting a first matching result;

[0009] If there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output.

[0010] Furthermore, generating a document summary corresponding to each document based on each document characteristic value includes:

[0011] According to a preset priority, selecting valid information with the highest priority among the document characteristic values ​​as the document summary;

[0012] Alternatively, information in the document characteristic value is selected according to a preset rule, and the information is merged to obtain the document summary.

[0013] The embodiment of the present invention provides two summary generation methods. First, a priority-based summary generation method is used to simplify the document summary generation process and improve generation efficiency. Second, when dealing with situations where information is missing, an information merging-based method is used to improve the integrity of the generated summary information, thereby improving the accuracy of subsequent matching. By adopting two generation methods for different situations, more dimensional situations are taken into consideration, thereby improving the accuracy of subsequent matching.

[0014] Furthermore, the bills and the documents are matched one by one according to the characteristic values ​​of the bills, the characteristic values ​​of the documents and the summaries of the documents, and a first matching result is output, which is specifically:

[0015] calculating whether each bill and each document satisfies a first matching condition based on the bill characteristic value, the document characteristic value, and the document summary, and outputting a first matching result if the first matching condition is satisfied;

[0016] If the first matching condition is not met, then calculate whether each bill and each document meets the second matching condition based on the bill feature value, the document feature value and the document summary, and output the first matching result if the second matching condition is met.

[0017] The embodiment of the present invention provides two matching conditions, so as to match all bills and documents as much as possible and output accurate and complete matching results.

[0018] Furthermore, the first matching condition is that the first matching factors all meet the preset requirements.

[0019] Among them, the first matching factors all meet preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the bill summary being equal to the document summary.

[0020] The embodiment of the present invention improves the accuracy of matching financial bills and documents by preferentially adopting the first matching condition with strict matching requirements.

[0021] Furthermore, the second matching condition is that the second matching factors all meet the preset requirements.

[0022] Among them, the second matching factors all meet the preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the similarity between the bill summary and the document summary meeting the preset conditions.

[0023] The embodiment of the present invention further adopts a second matching condition with relatively loose matching requirements to ensure the matching rate of financial bills and documents.

[0024] Furthermore, if there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output, specifically:

[0025] For unmatched bills and unmatched documents, extract the amount of each bill and the amount of each document according to the corresponding bill feature value and the corresponding document feature value;

[0026] Based on the amount of each bill and each document, calculate the amount difference according to the preset rules and determine whether the amount difference meets the preset requirements;

[0027] The bills and documents corresponding to the amount difference that meets the preset requirements are matched and output as the second matching result.

[0028] The embodiment of the present invention further improves the matching rate of financial bills and documents by using fuzzy matching for unmatched bills and documents.

[0029] Furthermore, the determination of whether the amount difference satisfies a preset condition is as follows:

[0030] If the absolute value of the difference between the document amount of a single document and a certain invoice amount is less than a preset threshold, the absolute value is taken as the amount difference, and it is determined that the amount difference meets the preset condition;

[0031] If the absolute value of the difference between the document amount of a single document and a certain invoice amount is not less than a preset threshold, the remaining documents are combined according to the preset rules to obtain a document combination, and the total document amount of each document combination is calculated;

[0032] The remaining bill amount is deducted from the total amount of each document and the absolute value is taken. The absolute value is taken as the amount difference. If the amount difference is less than the preset threshold, it is determined that the amount difference meets the preset condition.

[0033] The embodiment of the present invention determines whether the amount difference of each matching result meets the preset conditions, and further considers the document combination when a matching result that meets the conditions cannot be obtained for a single document, thereby improving the rigor of the matching result.

[0034] The embodiment of the present invention obtains bill characteristic values ​​and document characteristic values ​​by acquiring financial bill information and document information, providing a unified data format for subsequent fully automatic matching of financial bills and documents, facilitating subsequent data processing; by generating a document summary, data support is provided for subsequent matching actions; one-to-one matching is performed in advance based on the bill characteristic values, document characteristic values, and document summary, and matching is performed from multiple dimensions to ensure the accuracy of the first matching result; if there are unmatched bills and unmatched documents, fuzzy matching is performed, and by relaxing the matching conditions, it is ensured that bills with ambiguous information can also obtain matching results. Compared with the existing technology, the present invention improves the accuracy of fully automatic matching of financial bills and documents under fuzzy conditions.

[0035] In a second aspect, an embodiment of the present invention provides a fully automatic matching device for financial bills and documents under fuzzy conditions, comprising an information acquisition module, a summary generation module, a first matching module, and a fuzzy matching module, wherein:

[0036] The information acquisition module is used to obtain financial bill information and document information, and perform feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document;

[0037] The summary generation module is used to generate a document summary corresponding to each document according to each document feature value;

[0038] The first matching module is configured to match each of the bills with each of the documents according to each of the bill feature values, each of the document feature values, and each of the document summaries, and output a first matching result;

[0039] The fuzzy matching module is used to perform fuzzy matching according to the corresponding bill feature value and the corresponding document feature value when there are unmatched bills and unmatched documents, and output a second matching result.

[0040] The embodiment of the present invention obtains financial bill information and document information through an information acquisition module, and extracts bill feature values ​​and document feature values, thereby providing data preparation for subsequent fully automatic matching of financial bills and documents; generates a document summary through a summary generation module, thereby providing more dimensional data support for subsequent fully automatic matching; then, firstly, through a first matching module with strict matching requirements, the accuracy of fully automatic matching of financial bills and documents is improved; finally, fuzzy matching is performed on unmatched bills and documents through a fuzzy matching module with more relaxed matching requirements, thereby improving the matching rate of financial bills and documents, and further improving the accuracy of fully automatic matching of financial bills and documents under fuzzy conditions.

[0041] In a third aspect, an embodiment of the present invention provides a terminal device, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0042] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of the above.

[0043] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device / apparatus where the computer-readable storage medium is located is controlled to execute the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of the above items.

[0044] The above description is only an overview of the technical solutions of the embodiments of the present invention. In order to more clearly understand the technical means of the embodiments of the present invention, they can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiments of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of a method for fully automatic matching of financial bills and documents under fuzzy conditions provided by an embodiment of the present invention;

[0046] Figure 2 This is a structural diagram of a device for fully automatic matching of financial bills and documents under fuzzy conditions provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] Example 1:

[0049] like Figure 1 As shown, a method for fully automatic matching of financial bills and documents under fuzzy conditions provided by an embodiment of the present invention includes the following steps:

[0050] S11, obtaining financial bill information and document information, and performing feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document;

[0051] S12, generating a document summary corresponding to each document according to each document characteristic value;

[0052] S13, matching each of the bills with each of the documents based on the characteristic values ​​of each bill, the characteristic values ​​of each document, and the summary of each document, and outputting a first matching result;

[0053] S14: If there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output.

[0054] Preferably, the embodiment of the present invention exports financial bill information from a financial system, wherein the bill information includes but is not limited to bill date, income and expenditure category, bill amount, bill summary, and voucher number.

[0055] Furthermore, the embodiment of the present invention obtains documents from the financial system. There are two ways to obtain the documents: the first is to download a text format file from the financial system, generally in PDF format. After the document is directly downloaded from the system, no further text processing is required;

[0056] The second method is to download a non-text file from the financial system, usually in image format. After downloading the original document, it is converted into a text file through OCR recognition. The converted text format is usually PDF format.

[0057] Furthermore, feature extraction is performed on the financial bill information and the document information respectively to obtain a bill feature value corresponding to each bill and a document feature value corresponding to each document.

[0058] Preferably, the bill characteristic value includes bill number, bill date, income and expenditure category, bill amount, bill summary and voucher number;

[0059] The document characteristic values ​​include document number, document date, document amount, payee account and name, payee account and name, purpose, remarks, postscript and business type.

[0060] As a first aspect of the first embodiment of the present invention, generating a document summary corresponding to each document based on each document characteristic value includes:

[0061] According to a preset priority, selecting valid information with the highest priority among the document characteristic values ​​as the document summary;

[0062] Alternatively, information in the document characteristic value is selected according to a preset rule, and the information is merged to obtain the document summary.

[0063] Optionally, the priority order provided by the embodiment of the present invention is: purpose > remarks > postscript > other information. The method for judging the above-mentioned valid information is to check each billing information in turn according to the above-mentioned priority. If the checked information is not an empty value, the information is considered valid.

[0064] Optionally, in an embodiment of the present invention, information from the document characteristic values ​​is selected according to a preset rule and merged to obtain the document summary, wherein the preset rule is to select document information with non-null values ​​for merging. For example, in one embodiment, if the purpose, remarks, and postscript in the document information are all non-null values, then the document summary is equal to the purpose + remarks + postscript.

[0065] It should be noted that the embodiment of the present invention preferably adopts a priority-based document summary generation method, and when a large proportion of null values ​​in the corresponding document information is present, an information merging method is used to generate a summary.

[0066] As a second aspect of the first embodiment of the present invention, the bills and the documents are matched one by one based on the bill characteristic values, the document characteristic values, and the document summaries, and a first matching result is output, specifically:

[0067] calculating whether each bill and each document satisfies a first matching condition based on the bill characteristic value, the document characteristic value, and the document summary, and outputting a first matching result if the first matching condition is satisfied;

[0068] If the first matching condition is not met, then calculate whether each bill and each document meets the second matching condition based on the bill feature value, the document feature value and the document summary, and output the first matching result if the second matching condition is met.

[0069] Preferably, the first matching result output by the embodiment of the present invention includes, in addition to the successfully matched bill number and document number, a corresponding risk reminder.

[0070] It should be noted that the embodiment of the present invention gives priority to using the first matching condition. If there are still unmatched bills and documents after matching using the first matching condition, the second matching condition will be used.

[0071] As a third aspect of the first embodiment of the present invention, the first matching condition is that the first matching factors all meet the preset requirements.

[0072] Among them, the first matching factors all meet preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the bill summary being equal to the document summary.

[0073] As a fourth aspect of the first embodiment of the present invention, the second matching condition is that the second matching factors all meet the preset requirements.

[0074] Among them, the second matching factors all meet the preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the similarity between the bill summary and the document summary meeting the preset conditions.

[0075] Preferably, the embodiment of the present invention uses LCS similarity to calculate the similarity between the bill summary and the document summary, and the similarity meets the preset condition that the similarity is greater than or equal to a preset similarity threshold, and the similarity threshold is preferably set between 0.6 and 1.0.

[0076] It should be noted that the similarity threshold can be set according to actual matching requirements.

[0077] As a fifth aspect of the first embodiment of the present invention, if there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output, specifically:

[0078] For unmatched bills and unmatched documents, extract the amount of each bill and the amount of each document according to the corresponding bill feature value and the corresponding document feature value;

[0079] Based on the amount of each bill and each document, calculate the amount difference according to the preset rules and determine whether the amount difference meets the preset requirements;

[0080] The bills and documents corresponding to the amount difference that meets the preset requirements are matched and output as the second matching result.

[0081] Preferably, the second matching result output by the embodiment of the present invention includes, in addition to the successfully matched bill number and document number, a corresponding risk reminder.

[0082] As a sixth aspect of the first embodiment of the present invention, the determination of whether the amount difference satisfies a preset condition is specifically as follows:

[0083] If the absolute value of the difference between the document amount of a single document and a certain invoice amount is less than a preset threshold, the absolute value is taken as the amount difference, and it is determined that the amount difference meets the preset condition;

[0084] If the absolute value of the difference between the document amount of a single document and a certain invoice amount is not less than a preset threshold, the remaining documents are combined according to the preset rules to obtain a document combination, and the total document amount of each document combination is calculated;

[0085] The remaining bill amount is deducted from the total amount of each document and the absolute value is taken. The absolute value is taken as the amount difference. If the amount difference is less than the preset threshold, it is determined that the amount difference meets the preset condition.

[0086] Preferably, in one embodiment, the selected preset threshold is 0.01 yuan.

[0087] Preferably, after completing the matching of the financial bills and the documents, the embodiment of the present invention further includes displaying the matching results. The display process is specifically as follows:

[0088] The matching invoice number, document number, and corresponding reminder are placed in the final invoice. The invoice information corresponding to each document is calculated, and key information is annotated in the document. Typically, the key information is the voucher number or the invoice location. The annotated documents are sorted according to pre-set rules, deleted, and saved to obtain the final document file. This final document file is saved for subsequent review by users.

[0089] The embodiment of the present invention obtains bill characteristic values ​​and document characteristic values ​​by acquiring financial bill information and document information, providing a unified data format for subsequent fully automatic matching of financial bills and documents, facilitating subsequent data processing; by generating a document summary, data support is provided for subsequent matching actions; one-to-one matching is performed in advance based on the bill characteristic values, document characteristic values, and document summary, and matching is performed from multiple dimensions to ensure the accuracy of the first matching result; if there are unmatched bills and unmatched documents, fuzzy matching is performed, and by relaxing the matching conditions, it is ensured that bills with ambiguous information can also obtain matching results. Compared with the existing technology, the present invention improves the accuracy of fully automatic matching of financial bills and documents under fuzzy conditions.

[0090] Example 2:

[0091] like Figure 2 As shown, this embodiment provides a fully automatic matching device for financial bills and documents under fuzzy conditions, including an information acquisition module 001, a summary generation module 002, a first matching module 003 and a fuzzy matching module 004, wherein:

[0092] The information acquisition module 001 is used to obtain financial bill information and document information, and perform feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document;

[0093] The summary generation module is used to generate a document summary corresponding to each document according to each document feature value at 002;

[0094] The first matching module 003 is configured to match each bill with each document one by one according to each bill feature value, each document feature value, and each document summary, and output a first matching result;

[0095] The fuzzy matching module 004 is used to perform fuzzy matching according to the corresponding bill feature value and the corresponding document feature value when there are unmatched bills and unmatched documents, and output a second matching result.

[0096] In this embodiment, the summary generation module is used to generate a document summary corresponding to each document according to each document feature value, including:

[0097] According to a preset priority, selecting valid information with the highest priority among the document characteristic values ​​as the document summary;

[0098] Alternatively, information in the document characteristic value is selected according to a preset rule, and the information is merged to obtain the document summary.

[0099] In this embodiment, the first matching module 003 is used to match each bill with each document one by one according to each bill feature value, each document feature value, and each document summary, and output a first matching result, specifically:

[0100] The first matching module 003 calculates whether each bill and each document meets a first matching condition based on the bill feature value, the document feature value, and the document summary, and outputs a first matching result if the first matching condition is met;

[0101] If the first matching condition is not met, the first matching module 003 calculates whether each bill and each document meets the second matching condition based on the bill feature value, the document feature value and the document summary, and outputs the first matching result if the second matching condition is met.

[0102] In this embodiment, the fuzzy matching module 004 is used to perform fuzzy matching based on the corresponding bill feature value and the corresponding document feature value when there are unmatched bills and unmatched documents, and output a second matching result, specifically:

[0103] The fuzzy matching module 004 extracts the amount of each bill and the amount of each document from the unmatched bills and the unmatched documents according to the corresponding bill feature values ​​and the corresponding document feature values;

[0104] Based on the amount of each bill and each document, calculate the amount difference according to the preset rules and determine whether the amount difference meets the preset requirements;

[0105] The bills and documents corresponding to the amount difference that meets the preset requirements are matched and output as the second matching result.

[0106] For more detailed working principles and process steps of this embodiment, please refer to, but not limited to, the relevant records of Embodiment 1.

[0107] The embodiment of the present invention obtains financial bill information and document information through the information acquisition module 001, and extracts bill feature values ​​and document feature values ​​to provide data preparation for the subsequent fully automatic matching of financial bills and documents; generates a document summary through the summary generation module 002 to provide more dimensional data support for the subsequent fully automatic matching; then, first uses the first matching module 003 with strict matching requirements to improve the accuracy of the fully automatic matching of financial bills and documents; finally, fuzzy matching is performed on unmatched bills and documents through the fuzzy matching module 004 with more relaxed matching requirements, thereby improving the matching rate of financial bills and documents, and further improving the accuracy of the fully automatic matching of financial bills and documents under fuzzy conditions.

[0108] Example 3:

[0109] This embodiment provides a terminal device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0110] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of the above.

[0111] Example 4:

[0112] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device / apparatus where the computer-readable storage medium is located is controlled to execute the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of the above items.

[0113] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-monitorable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0114] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for fully automatic matching of financial bills and documents under fuzzy conditions, characterized in that: include: Obtain financial bill information and document information, and perform feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document; Generating a document summary corresponding to each document according to each document characteristic value; Matching each of the bills with each of the documents according to the characteristic values ​​of each bill, the characteristic values ​​of each document, and the summary of each document, and outputting a first matching result; If there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output.

2. A method for fully automatic matching of financial bills and documents under fuzzy conditions as claimed in claim 1, characterized in that: Generating a document summary corresponding to each document according to each document characteristic value includes: According to a preset priority, selecting valid information with the highest priority among the document characteristic values ​​as the document summary; Alternatively, information in the document characteristic value is selected according to a preset rule, and the information is merged to obtain the document summary.

3. The method for fully automatic matching of financial bills and documents under fuzzy conditions as claimed in claim 1, characterized in that: According to each of the bill characteristic values, each of the document characteristic values ​​and each of the document summaries, each of the bills and each of the documents is matched one by one, and a first matching result is output, specifically: calculating whether each bill and each document satisfies a first matching condition based on the bill characteristic value, the document characteristic value, and the document summary, and outputting a first matching result if the first matching condition is satisfied; If the first matching condition is not met, then calculate whether each bill and each document meets the second matching condition based on the bill feature value, the document feature value and the document summary, and output the first matching result if the second matching condition is met.

4. A method for fully automatic matching of financial bills and documents under fuzzy conditions as claimed in claim 3, characterized in that: The first matching condition is that the first matching factors all meet the preset requirements. Among them, the first matching factors all meet preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the bill summary being equal to the document summary.

5. A method for fully automatic matching of financial bills and documents under fuzzy conditions as claimed in claim 3, characterized in that: The second matching condition is that the second matching factors all meet the preset requirements. Among them, the second matching factors all meet the preset requirements, including the amount difference meeting the preset requirements, the income and expenditure categories matching, the bill date being equal to the document date, and the similarity between the bill summary and the document summary meeting the preset conditions.

6. A method for fully automatic matching of financial bills and documents under fuzzy conditions as claimed in claim 1, characterized in that: If there are unmatched bills and unmatched documents, fuzzy matching is performed based on the corresponding bill feature value and the corresponding document feature value, and a second matching result is output, specifically: For unmatched bills and unmatched documents, extract the amount of each bill and the amount of each document according to the corresponding bill feature value and the corresponding document feature value; Based on the amount of each bill and each document, calculate the amount difference according to the preset rules and determine whether the amount difference meets the preset requirements; The bills and documents corresponding to the amount difference that meets the preset requirements are matched and output as the second matching result.

7. A method for fully automatic matching of financial bills and documents under fuzzy conditions according to any one of claims 4 to 6, characterized in that: The determination of whether the amount difference meets the preset conditions is specifically as follows: If the absolute value of the difference between the document amount of a single document and a certain invoice amount is less than a preset threshold, the absolute value is taken as the amount difference, and it is determined that the amount difference meets the preset condition; If the absolute value of the difference between the document amount of a single document and a certain invoice amount is not less than a preset threshold, the remaining documents are combined according to the preset rules to obtain a document combination, and the total document amount of each document combination is calculated; The remaining bill amount is deducted from the total amount of each document and the absolute value is taken. The absolute value is taken as the amount difference. If the amount difference is less than the preset threshold, it is determined that the amount difference meets the preset condition.

8. A fully automatic matching device for financial bills and documents under fuzzy conditions, characterized in that: It includes an information acquisition module, a summary generation module, a first matching module and a fuzzy matching module, wherein: The information acquisition module is used to obtain financial bill information and document information, and perform feature extraction on the financial bill information and the document information respectively to obtain bill feature values ​​corresponding to each bill and document feature values ​​corresponding to each document; The summary generation module is used to generate a document summary corresponding to each document according to each document feature value; The first matching module is configured to match each of the bills with each of the documents one by one according to each of the bill feature values, each of the document feature values, and each of the document summaries, and output a first matching result; The fuzzy matching module is used to perform fuzzy matching according to the corresponding bill feature value and the corresponding document feature value when there are unmatched bills and unmatched documents, and output a second matching result.

9. A terminal device, characterized in that: include: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein, when the computer program is running, the device / apparatus where the computer-readable storage medium is located is controlled to execute the fully automatic matching method of financial bills and documents under fuzzy conditions as described in any one of claims 1 to 7.

Images