Business background verification method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of artificial intelligence, and discloses a business background verification method and device, electronic equipment and a storage medium, the method comprising the following steps: acquiring original background information corresponding to a business matter; extracting business indexes under each verification dimension from the original background information; verifying the business indexes by using a verification model based on verification rules corresponding to the business indexes, to obtain verification results corresponding to the business indexes; and determining a background verification result corresponding to the business matter based on the verification results under each verification dimension. By implementing the method, the accuracy and efficiency of business background verification of the business matter can be improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to business background verification methods, devices, electronic equipment, and storage media. Background Technology

[0002] In the construction industry supply chain, general contractors, specialized subcontractors, and building material suppliers frequently interact, resulting in various business transactions, such as contracts for the purchase of building materials. Currently, the business background verification is typically conducted based on the original background information corresponding to these transactions, such as contracts and settlement statements, using text comparison tools or rule engines.

[0003] However, due to the lack of context-specific reasoning capabilities in text comparison tools or rule engines, and their limited ability to process unstructured information such as scanned documents, images, seals, and signatures, the accuracy of business background verification for business matters is not high. Summary of the Invention

[0004] This invention provides a business background verification method, apparatus, electronic device, and storage medium to solve the problem of low accuracy in verifying the business background of business matters.

[0005] In a first aspect, the present invention provides a business background verification method, the method comprising: obtaining original background information corresponding to a business matter; extracting business indicators under each verification dimension from the original background information; verifying the business indicators using a verification model based on the verification rules corresponding to the business indicators, and obtaining the verification results corresponding to the business indicators; and determining the background verification result corresponding to the business matter based on the verification results under each verification dimension.

[0006] The business background verification method provided in this embodiment extracts business indicators under various verification dimensions from the original background information corresponding to the business matter. Then, it uses a verification model to verify the business indicators according to the verification rules corresponding to the business indicators to obtain verification results. Finally, it determines the background verification result corresponding to the business matter based on the verification results under each verification dimension. Since the verification model utilizes its reasoning ability and the verification rules corresponding to the business indicators to verify the business indicators, it can solve the problem of low accuracy in background verification of business matters caused by text comparison tools or rule engines focusing only on character matching and relying on preset rules. Therefore, it improves the efficiency of background verification for business matters.

[0007] In one optional implementation, the verification model is used to verify the business indicators based on the verification rules corresponding to the business indicators, and the verification results corresponding to the business indicators are obtained. This includes: searching the verification rule base based on the business indicators to determine the first-level verification rules corresponding to the business indicators, wherein the verification rule base is used to store the mapping relationship between the business indicators and the first-level verification rules; and verifying the business indicators based on the first-level verification rules corresponding to the business indicators using the verification model to obtain the verification results corresponding to the business indicators.

[0008] By searching the verification rule base based on business metrics, the first-level verification rules corresponding to the business metrics can be retrieved quickly and accurately. Then, the verification model is used to perform preliminary verification of the business metrics according to the first-level verification rules. This can filter out obviously non-compliant business metrics in advance, reduce the amount of computation for subsequent verification, and improve the efficiency and accuracy of the overall verification process.

[0009] In an optional implementation, the method further includes: if the verification result indicates that the business indicator has failed verification, then searching the verification rule base based on the business indicator to determine the second-level verification rule corresponding to the business indicator, wherein the verification granularity of the second-level verification rule is smaller than that of the first-level verification rule; and verifying the business indicator using the verification model based on the second-level verification rule corresponding to the business indicator to obtain a new verification result corresponding to the business indicator.

[0010] If the verification result indicates that the business indicator has failed the verification, it means that the business indicator needs to be further reviewed. Since the verification granularity corresponding to the second-level verification rule is smaller than that corresponding to the first-level verification rule, the verification model is used to verify the business indicator according to the second-level verification rule in order to further explore the specific problems of the business indicator. In this way, more accurate new verification results can be obtained.

[0011] In one optional implementation, business indicators under each verification dimension are extracted from the original background information, including: obtaining target prompt words corresponding to each extraction model; for each extraction model, using the target prompt words to guide the extraction model to extract the initial business indicators under each verification dimension in the original background information; for each verification dimension, comparing the initial business indicators extracted by all extraction models corresponding to the verification dimension to obtain the business indicators under the verification dimension, thus obtaining the business indicators corresponding to each verification dimension.

[0012] By configuring corresponding target prompts for each extraction model, the model can be guided to extract business indicators under various verification dimensions from the original background information in a preset manner, reducing extraction bias. At the same time, by comparing and fusing the initial business indicators of multiple extraction models, erroneous or abnormal extraction results can be effectively eliminated, improving the accuracy of the final business indicators.

[0013] In one optional implementation, each extraction model includes a first extraction model and a second extraction model; the initial business indicators extracted by all extraction models corresponding to the verification dimension are compared to obtain the business indicators under the verification dimension, including: comparing the first initial business indicator corresponding to the first extraction model with the second initial business indicator corresponding to the second extraction model to obtain a comparison result; if the comparison result indicates that the first initial business indicator and the second initial business indicator are the same, then the first initial business indicator or the second initial business indicator is determined as the business indicator under the verification dimension; if the comparison result indicates that the first initial business indicator and the second initial business indicator are different, then a new first initial business indicator is re-extracted using the first extraction model, and the business indicators under the verification dimension are determined based on the new first initial business indicator and the second initial business indicator.

[0014] The first initial business indicator extracted by the first extraction model is compared with the second initial business indicator extracted by the second extraction model. If the first initial business indicator and the second initial business indicator are different, it indicates that the extraction result of at least one extraction model may have an error. In this case, re-extracting can effectively reduce the impact of a single extraction error on the final business indicator, making the business indicators under the final determined verification dimension more accurate.

[0015] In one optional implementation, the process of obtaining the target prompt word corresponding to each extraction model includes: obtaining multiple training background information, the label business indicators corresponding to each training background information, and the initial prompt word corresponding to the extraction model; using the initial prompt word to guide the extraction model to extract the training business indicators from each training background information; comparing the label business indicators and training business indicators corresponding to each training background information to obtain the accuracy corresponding to each training background information; if there is an accuracy less than a preset threshold, the initial prompt word is adjusted to obtain the target prompt word; if all accuracy rates are greater than or equal to the preset threshold, the initial prompt word is determined as the target prompt word.

[0016] By comparing the training business metrics generated by the extraction model based on the initial prompt words with the tag business metrics, the guiding effect of the initial prompt words can be quantified. When the accuracy does not meet the requirements, i.e., the accuracy is less than the preset threshold, the initial prompt words are adjusted to obtain more accurate target prompt words, which can further improve the accuracy of business metric extraction.

[0017] In one optional implementation, the background verification result corresponding to the business item is determined based on the verification results under each verification dimension, including: if all verification results indicate that the business indicator corresponding to the verification result has passed the verification, then a background verification result indicating that the business item has passed the verification is obtained; if any verification result indicates that the business indicator corresponding to the verification result has failed the verification, then a background verification result indicating that the business item has failed the verification is obtained.

[0018] By adopting the logic that all business indicators must pass verification before a business item is deemed to have passed verification, it is possible to ensure that business items meet the requirements in all verification dimensions such as authenticity, validity, and reasonableness, thereby avoiding misjudgments caused by the omission of a single business indicator and improving the reliability of background verification results.

[0019] Secondly, the present invention provides a business background verification device, which includes: a first acquisition module for acquiring original background information corresponding to a business matter; a first extraction module for extracting business indicators under each verification dimension from the original background information; a first verification module for verifying the business indicators using a verification model based on the verification rules corresponding to the business indicators, and obtaining the verification results corresponding to the business indicators; and a first determination module for determining the background verification result corresponding to the business matter based on the verification results under each verification dimension.

[0020] Thirdly, the present invention provides an electronic device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the business background verification method described in the first aspect or any corresponding embodiment thereof.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the business background verification method described in the first aspect or any corresponding embodiment thereof.

[0022] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the business background verification method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of the first type of business background verification method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a display page according to an embodiment of the present invention; Figure 4 This is a schematic diagram of an upload page according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a background verification report according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the second process of the business background verification method according to an embodiment of the present invention; Figure 7 This is a flowchart illustrating a business background verification method according to an embodiment of the present invention; Figure 8 This is a structural block diagram of a business background verification device according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0028] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0029] In the construction industry supply chain, frequent interactions between general contractors, specialized subcontractors, and building material suppliers result in various business transactions, such as contracts for the purchase of building materials. Typically, background verification is required to determine the authenticity, reasonableness, and validity of these transactions.

[0030] However, since business background verification in the construction industry requires combining professional knowledge such as construction specifications and building material characteristics, manual verification of business backgrounds relies on the engineering experience of the verifiers. Different verifiers have different engineering experience, and consequently, the standards for verifying business backgrounds also differ. Therefore, manual verification of business backgrounds is prone to an imbalance between efficiency and verification standards.

[0031] Based on the above problems, using general tools such as text comparison tools or rule engines to verify the business background of business matters can, to some extent, solve the problem of the imbalance between efficiency and standards caused by manual business background verification. However, text comparison tools cannot recognize the differences in professional terminology in the construction industry and can only achieve general character matching, lacking industry-specific reasoning capabilities. Rule engines cannot cover industry-specific scenarios and cannot be linked to construction progress data, thus easily generating false risk reports. Furthermore, text comparison tools or rule engines have limited processing capabilities for unstructured information such as scanned documents, images, seals, or signatures, resulting in low accuracy in verifying the business background of business matters.

[0032] To address the aforementioned shortcomings, this application utilizes the unstructured information processing capabilities, semantic understanding capabilities, and logical reasoning capabilities of large-scale language models to verify the business background of business matters, thereby improving the efficiency and accuracy of business background verification.

[0033] As an optional application scenario of this invention, such as Figure 1 As shown, application 101 is installed in terminal device 110, and user 130 can interact with application 101 through terminal device 110 and / or access device of terminal device 110.

[0034] For example, application 101 can be any application that can provide business context verification for related services. Figure 1 In the application scenario shown, if application 101 is active, the terminal device 110 can display the interface 102 of application 101. The interface 102 may include various pages that application 101 can provide, such as interactive pages, display pages, upload pages, etc.

[0035] In some embodiments, terminal device 110 is communicatively connected to server 120 to provide services to application 101. Terminal device 110 may be a mobile terminal, fixed terminal, or portable terminal, etc., including but not limited to mobile phones, desktop computers, laptop computers, multimedia tablets, e-book devices, gaming devices, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some embodiments, terminal device 110 may also support any type of interface, and server 120 may be various types of computing systems or servers capable of providing computing power, including but not limited to mainframes, edge computing nodes, computing devices in cloud environments, etc.

[0036] It should be noted that, Figure 1 This is merely an example of an application scenario and does not limit the scope of protection of this invention.

[0037] The embodiments of the present invention will now be described with reference to the accompanying drawings. It should be understood that the pages shown in the drawings are merely examples, and various page designs are possible in practice. The various graphic elements on the page may have different arrangements and different visual representations; one or more elements may be omitted or replaced, and one or more other elements may also be present, without any limitation in the embodiments of the present invention. Furthermore, the embodiments described below primarily pertain to terminal device 110. It should be understood that the actions described relative to terminal device 110 can be performed by application 101 on terminal device 110, or can be performed by application 101 in conjunction with its server (e.g., server 120).

[0038] According to an embodiment of the present invention, a business background verification method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0039] This embodiment provides a business background verification method that can be used in terminal devices. Figure 2 This is a flowchart of a business background verification method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain the original background information corresponding to the business matter.

[0040] Here, a business transaction can refer to a specific event arising from service interaction between service providers and consumers in a particular business scenario, requiring a series of processing steps to complete. Correspondingly, the original background information can be a series of original documents generated throughout the entire lifecycle of the business transaction, capable of proving its occurrence, development, and outcome.

[0041] For example, in the construction industry, construction companies, as the demand side, need to purchase building materials from building material suppliers, who are the suppliers. In this scenario, the specific business transaction arising from the purchase of building materials between the construction company and the building material supplier is the purchase of building materials. Accordingly, in the process of purchasing building materials, a series of original background information is generated and accumulated to prove the occurrence and flow of the business, including but not limited to: purchase contracts, delivery receipts, settlement statements, and corresponding VAT invoices and other original documents.

[0042] For example, in the field of corporate procurement, the administrative department needs to purchase office supplies from suppliers. Therefore, a procurement transaction arises between the administrative department and the supplier due to the purchase of materials. Accordingly, the procurement process will generate purchase requisition forms, purchase contracts, delivery receipts, and invoices corresponding to the payment, etc.

[0043] In the construction industry, construction companies typically use project management systems. The procurement module within these systems encompasses comprehensive management of all transactions with building material suppliers. Contracts, delivery notes, settlement statements, and invoices constitute the foundational data for this module. After authorization by the construction company, the application's business background verification system (i.e....) Figure 1 The data acquisition module in application 101) shown can access the procurement module of the project management system to obtain the original background information of contracts and their corresponding delivery notes, settlement statements, invoices, etc., and treat the transactions corresponding to the contracts as business items. Then, the business items and their corresponding original background information can be displayed as follows: Figure 3 The page shown is displayed as shown. (As shown on the display page.) Figure 3 As shown, the display page can specifically display information such as the business item number, business item type, purchaser name, supplier name, business item, and operation details. The business item number can be generated by the business background verification system of this application based on built-in coding rules, and the business item type can be determined based on the business item.

[0044] Of course, users can click Figure 3 The details control shown displays the original background information of the business transaction corresponding to the details control. Additionally, users can click... Figure 3 The business background verification control shown can be responded to by the user's click operation on the terminal device, and the background verification of the business item corresponding to the business background verification control can be started.

[0045] Of course, the business background verification system of this application can also provide, for example... Figure 4The upload page shown allows users to upload original background information related to their business transactions. For example, in the construction industry, the original background information for procuring building materials can include, but is not limited to, detailed registration information from the China Securities Depository and Clearing Corporation (CSDC), contract information, settlement statement information, and invoice information. Users can also enter keywords related to business background verification. After uploading, users can click [the upload button]. Figure 4 In the context of business background verification controls, the terminal device responds to the user's click and begins background verification of the business items corresponding to the control. For example... Figure 4 As shown, each information upload area has corresponding prompts. For example, the first upload area for detailed registration information has a prompt such as "Click or drag a file to this area for upload recognition. Multiple files can be uploaded, and zip and pdf formats are supported." The second upload area for contract information has a prompt such as "File upload. Multiple files can be uploaded, and pdf, jpg, peg, and png formats are supported." The third upload area for settlement statement information has a prompt such as "File upload. Multiple files can be uploaded, and pdf, jpg, peg, and png formats are supported." The fourth upload area for invoice information has a prompt such as "File upload. Multiple files can be uploaded, and pdf, jpg, peg, and png formats are supported."

[0046] Step S202: Extract business metrics for each verification dimension from the original background information.

[0047] The verification dimensions here can be authenticity, validity, and reasonableness. That is, this application can verify the business background of business matters from three dimensions: authenticity, validity, and reasonableness. Correspondingly, business indicators can be key information used to reflect the business matters under different verification dimensions such as authenticity, validity, and reasonableness when verifying the business background. This key information can be extracted from original background information such as contracts, settlement statements, and invoices. Moreover, each verification dimension can correspond to one or more business indicators.

[0048] Taking contract type as an example, under the dimension of authenticity, business indicators can include, but are not limited to, contract parties, transaction history information, procurement information, project information, performance date, performance category, performance unit price, performance quantity, performance amount, delivery address, authenticity of handling, approval process, unit price of goods, delivery quantity, etc. Under the dimension of reasonableness, business indicators can include, but are not limited to, overall time, overall amount, delivery category, delivery location, unit price of goods, delivery quantity, etc. Under the dimension of validity, business indicators can include, but are not limited to, debt transfer information and China Securities Depository and Clearing Corporation (CSDC) registration information, etc.

[0049] This process extracts business metrics for each verification dimension from the original background information. This can involve extracting key information required for each verification dimension, such as contract parties, transaction history, and procurement information. Alternatively, any feasible method can be used to extract the required verification information for each verification dimension from the original background information; this application does not impose any specific limitations on this. For example, business metrics for each verification dimension can be extracted from the original background information based on preset keyword matching rules, field position rules, regular expression matching rules, template matching rules, semantic rules, etc. Another example is using natural language processing technology to analyze the original background information to extract business metrics for each verification dimension. Yet another example is using pre-trained text classification models, multimodal recognition models, etc., to process the original background information and automatically extract business metrics for each verification dimension.

[0050] Step S203: Based on the verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the verification results corresponding to the business indicators.

[0051] As a concrete example, verification rules can be predefined rules used to verify business metrics. Each business metric can have a corresponding verification rule. For example, when the business metric is the contract subject, the verification rule can be used to verify whether the contract subject is genuine; when the business metric is transaction history information, the verification rule can be used to verify whether the transaction history information corresponding to the business item is genuine; when the business metric is procurement information, the verification rule can be used to verify whether the procurement information is genuine; when the business metric is total time, the verification rule can be used to verify the logical reasonableness of the total time; when the business metric is delivery location, the verification rule can be used to verify whether the delivery location is reasonable, and so on.

[0052] Here, the logical reasoning capabilities of the verification model can be used to verify business metrics, resulting in relatively accurate verification results. The verification model can be trained based on a large language model architecture, a machine learning model architecture, or even a combination of multiple model architectures. There are no specific limitations here, as long as it can verify the business metrics.

[0053] Step S204: Based on the verification results under each verification dimension, determine the background verification results corresponding to the business item.

[0054] As a concrete example, the weights of each verification dimension can be determined based on their importance. Then, the weights and verification results of each dimension are weighted together to determine the background verification result for the business transaction. Alternatively, the verification results from each dimension can be input into a fusion model, allowing the model to infer the background verification result based on these results. The training architecture of the fusion model is similar to that of the verification model described earlier and will not be repeated here.

[0055] In addition, after obtaining the background verification results corresponding to the business matter, the background verification report corresponding to the business matter can be obtained by integrating the verification results under each verification dimension and the background verification results corresponding to the business matter.

[0056] As a specific example, such as Figure 5 As shown, the background verification report may include a first display area, a second display area, a third display area, a fourth display area, a fifth display area, and a sixth display area. Among them, The first display area is used to display basic information corresponding to the background verification report, such as contract name, purchaser name, supplier name, report number of the background verification report, report generation time, and the operator of the background verification report.

[0057] The second display area is used to show the background verification results. If the results are not largely consistent, it is recommended to reject them.

[0058] The third display area is used to show the various business indicators under the validity verification dimension, as well as the corresponding verification results, risk recommendations, and raw data for each business indicator. Specifically, the third display area can show the verification results, risk recommendations, and raw data for the business indicator "Asset Transferability," and the verification results, risk recommendations, and raw data for the business indicator "Resource Transferred." The risk recommendations for each business indicator can be generated using a verification model, or they can be manually set. Additionally, the raw data can be the specific content corresponding to the business indicator.

[0059] The fourth display area is used to display the various business indicators under the verification dimension of authenticity, as well as the corresponding verification results, risk suggestions, and raw data for each business indicator. Specifically, the fourth display area can display the verification results, risk suggestions, and raw data for the business indicator of contract subject (Party A), contract subject (Party B), procurement information (Party A), and procurement information (Party B).

[0060] The fifth display area can show the various business indicators under the verification dimension of reasonableness, as well as the corresponding verification results, risk suggestions, and raw data for each business indicator. Specifically, the fifth display area can show the verification results, risk suggestions, and raw data for the business indicator of total time, and the verification results, risk suggestions, and raw data for the business indicator of total amount.

[0061] The sixth display area shows invoice verification details, such as the invoice number, invoice code, invoice type, total price including tax, invoice date, amount excluding tax, verification code, and verification status for invoices that passed verification, as well as the invoice number, invoice code, invoice type, total price including tax, invoice date, amount excluding tax, verification code, and verification status for invoices that failed verification. The risk recommendations for each business indicator can be generated using a verification model or manually set. Additionally, the raw data can be the specific content corresponding to the business indicator.

[0062] The business background verification method provided in this embodiment extracts business indicators under various verification dimensions from the original background information corresponding to the business matter. Then, it uses a verification model to verify the business indicators according to the verification rules corresponding to the business indicators to obtain verification results. Finally, it determines the background verification result corresponding to the business matter based on the verification results under each verification dimension. Since the verification model utilizes its reasoning ability and the verification rules corresponding to the business indicators to verify the business indicators, it can solve the problem of low accuracy in background verification of business matters caused by text comparison tools or rule engines focusing only on character matching and relying on preset rules. Therefore, it improves the efficiency of background verification for business matters.

[0063] This embodiment provides a business background verification method that can be used in terminal devices. Figure 6 This is a flowchart of a business background verification method according to an embodiment of the present invention, such as... Figure 6 As shown, the process includes the following steps: Step S601: Obtain the original background information corresponding to the business transaction. For details, please refer to [link / reference]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0064] Step S602: Extract business metrics for each verification dimension from the original background information.

[0065] Specifically, step S602 includes: Step S6021: Obtain the target prompt words corresponding to each extraction model.

[0066] The target prompt word can be text information input to the extraction model and used to guide it in generating specific output results. Here, one extraction model can correspond to one target prompt word, and one target prompt word can include multiple sub-prompt words corresponding to different verification dimensions. The training architecture of the extraction model is similar to that of the verification model, and will not be repeated here.

[0067] As a specific example, the business background verification system of this application has a built-in prompt word library. This prompt word library is used to store the correspondence between the extraction model and the target prompt words. Therefore, the target prompt words corresponding to the extraction model can be obtained by searching the prompt word library based on the prompt model.

[0068] Step S6022: For each extraction model, use target prompts to guide the extraction model to extract the initial business indicators under each verification dimension in the original background information.

[0069] As mentioned earlier, a target prompt word includes multiple sub-prompt words corresponding to different verification dimensions. Therefore, for an extraction model, guided by its corresponding target prompt word, it can extract the initial business indicators for each verification dimension from the original background information. For example, guided by its corresponding target prompt word, the extraction model can extract the initial business indicators corresponding to the verification dimensions of authenticity, validity, and rationality.

[0070] Accordingly, after each extraction model extracts the initial business indicators for each verification dimension from the original background information, from the perspective of the verification dimension, each verification dimension corresponds to multiple initial business indicators extracted by the extraction models. For example, if there are three extraction models, namely the first extraction model, the second extraction model, and the third extraction model, then under the verification dimension of authenticity, there can be initial business indicators for the authenticity verification dimension extracted by the first extraction model, the second extraction model, and the third extraction model.

[0071] In addition, before each extraction model extracts the initial business indicators under each verification dimension from the original background information based on its corresponding target prompt words, optical character recognition (OCR) technology can be used to convert the original background information represented in the form of PDF or image into text information.

[0072] Step S6023: For each verification dimension, compare the initial business indicators extracted by all extraction models corresponding to the verification dimension to obtain the business indicators under the verification dimension, thereby obtaining the business indicators corresponding to each verification dimension.

[0073] In some optional implementations, under the same verification dimension, the initial business indicators extracted by each extraction model under that verification dimension can be compared, the frequency of occurrence of each initial business indicator can be counted, and the initial business indicator with the highest frequency of occurrence can be determined as the business indicator under that verification dimension.

[0074] For example, if the number of extraction models is set to 5, and among the initial business indicators extracted by the 5 extraction models, 3 initial business indicators have the same content, and the other 2 initial business indicators have the same content, then the one that appears more frequently (i.e., the 3 identical initial business indicators) is selected as the final business indicator.

[0075] By configuring corresponding target prompts for each extraction model, the model can be guided to extract business indicators under various verification dimensions from the original background information in a preset manner, reducing extraction bias. At the same time, by comparing and fusing the initial business indicators of multiple extraction models, erroneous or abnormal extraction results can be effectively eliminated, improving the accuracy of the final business indicators.

[0076] In some optional implementations, step S6023 above further includes: Step a1: Compare the first initial business indicator corresponding to the first extraction model with the second initial business indicator corresponding to the second extraction model to obtain the comparison result.

[0077] Step a2: If the comparison results indicate that the first initial business indicator and the second initial business indicator are the same, then the first initial business indicator or the second initial business indicator is determined as the business indicator under the verification dimension.

[0078] Step a3: If the comparison results indicate that the first initial business indicator and the second initial business indicator are different, then the first extraction model is used to re-extract the new first initial business indicator, and the business indicators under the verification dimension are determined based on the new first initial business indicator and the second initial business indicator.

[0079] The process of determining the business indicators under the verification dimension based on the new first initial business indicator and the second initial business indicator can be as follows: The new first initial business indicator and the second initial business indicator are compared to obtain a new comparison result. If the new comparison result indicates that the new first initial business indicator and the second initial business indicator are the same, then the new first initial business indicator or the second initial business indicator is used as the business indicator under the verification dimension. If the new comparison result indicates that the new first initial business indicator and the second initial business indicator are different, then the first extraction model can be used to extract the new first initial business indicator again, and the aforementioned process continues until the new first initial business indicator and the second initial business indicator are the same. Of course, if the aforementioned comparison rounds have reached the maximum number of comparison rounds, then the initial business indicator extracted by the first extraction model or the second extraction model can be used as the final business indicator.

[0080] The first initial business indicator extracted by the first extraction model is compared with the second initial business indicator extracted by the second extraction model. If the first initial business indicator and the second initial business indicator are different, it indicates that the extraction result of at least one extraction model may have an error. In this case, re-extracting can effectively reduce the impact of a single extraction error on the final business indicator, making the business indicators under the final determined verification dimension more accurate.

[0081] In some optional implementations, the process of obtaining the target prompt words corresponding to each extraction model includes: Step b1: Obtain multiple training background information, the corresponding label business indicators for each training background information, and extract the initial prompt words corresponding to the model.

[0082] Step b2: Use the initial prompt words to guide the extraction model to extract training business indicators from various training background information.

[0083] Step b3: Compare the label business metrics and training business metrics corresponding to each training background information to obtain the accuracy corresponding to each training background information.

[0084] Step b4: If there is an accuracy rate lower than a preset threshold, the initial prompt word is adjusted to obtain the target prompt word.

[0085] Step b5: If all accuracy rates are greater than or equal to the preset threshold, then the initial prompt word is determined as the target prompt word.

[0086] The training background information here can be anything like contracts, invoices, or settlement statements, as mentioned earlier; there are no specific limitations. Furthermore, each piece of training background information can correspond to one or more labeled business metrics. For example, a contract can correspond to either the labeled business metric of the contracting party (Party A) or the labeled business metric of the contracting party (Party B). Accordingly, there is a one-to-one correspondence between the training business metrics and the labeled business metrics.

[0087] As a specific example, taking a training background information as an example, if the training background information corresponds to multiple label business indicators, the process of determining the accuracy corresponding to the training background information can be as follows: compare each training business indicator with the label business indicator to determine whether the training business indicator and the label business indicator are the same; count the number of training business indicators that are the same as the label business indicators; and then take the ratio of the number of training business indicators that are the same as the label business indicators to the total number of training business indicators or label business indicators as the accuracy, thereby obtaining the accuracy corresponding to the training background information.

[0088] The preset threshold here can be a pre-set threshold. This application does not limit the specific value of the preset threshold, which can be flexibly set according to the actual business scenario. For example, the preset threshold can be 0.9.

[0089] When an accuracy rate is less than a preset threshold, the specific process for adjusting the initial prompt word to obtain the target prompt word can be as follows: First, determine the target training background information where the accuracy rate is less than the preset threshold. Then, analyze the training business indicators corresponding to the target training background information to identify the reasons for the extraction errors in the training business indicators. Next, adjust the initial prompt word based on the reasons for the extraction errors to obtain candidate prompt words. Based on the candidate prompt words, continue to guide the extraction model corresponding to the candidate prompt words to extract new training business indicators from the target training background information. Then, compare the new training business indicators with the label business indicators to determine the new accuracy rate. If the new accuracy rate is greater than or equal to the preset threshold, then the candidate prompt word is determined as the target prompt word. If the new accuracy rate is less than the preset threshold, continue to adjust and iterate the candidate prompt words in the above manner until the new accuracy rate is greater than or equal to the preset threshold, thus obtaining the final target prompt word.

[0090] By comparing the training business metrics generated by the extraction model based on the initial prompt words with the tag business metrics, the guiding effect of the initial prompt words can be quantified. When the accuracy does not meet the requirements, i.e., the accuracy is less than the preset threshold, the initial prompt words are adjusted to obtain more accurate target prompt words, which can further improve the accuracy of business metric extraction.

[0091] Step S603: Based on the verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the verification results corresponding to the business indicators.

[0092] Specifically, step S603 includes: Step S6031: Determine the first-level verification rule corresponding to the business indicator based on the verification rule base retrieved from the business indicator. The verification rule base is used to store the mapping relationship between the business indicator and the first-level verification rule.

[0093] As a concrete example, each business metric corresponds to two different levels of verification rules: a first-level verification rule (basic verification rule) and a second-level verification rule (deep verification rule). Both the business metric and its corresponding two levels of verification rules are stored in a verification rule base. This base stores the mapping relationships between business metrics and the first and second-level verification rules. Subsequently, the verification rule base can be searched based on the business metric. If no verification result is found for a given business metric, the first-level verification rule corresponding to that metric is retrieved from the base. If a verification result is found for a given business metric, the second-level verification rule corresponding to that metric is retrieved from the base.

[0094] Taking the business indicator of procurement information under the verification dimension of authenticity as an example, its corresponding first-level verification rule can be: {Business indicator: Authenticity of procurement information (Party A); Verification dimension: Authenticity; Function: Verify the authenticity of procurement information and verify that the names of the buyer / Party A in the documents are consistent; Input requirements: htInput is the procurement information in the contract, jsdInput is the procurement information in the settlement statement, and fpInput is the procurement information in the invoice; Execution conditions: Retrieve all contract Party A, all settlement statement buyer, and all invoice buyer and output them; When all the above fields of all documents have no data, the verification result is Uncertainty; When all the above fields of all documents are completely consistent, they must be from the same company, and the verification result is true (basically consistent); When the above fields of any document are obviously not from the same company as other documents, the verification result is false (basically inconsistent); In other cases, when the above fields of the documents are not completely consistent, but all the contents are relatively similar, the verification result is plausible (there are uncertain items); Output format: Output in JSON format.} Strictly adhere to the following JSON format for output, ensuring consistent key values ​​and order, and simultaneously output explanatory text: Procurement Information Authenticity (Party A): {Contract Party A: xxx (array showing values ​​for each contract); Settlement Statement Buyer: xxx (array showing values ​​for each settlement statement); Invoice Buyer: xxx (array showing values ​​for each invoice) Verification Result: true / plausible / false; Explanation: Explanatory text} Verification Depth: First Level}

[0095] Step S6032: Based on the first-level verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the verification results corresponding to the business indicators.

[0096] As mentioned above, the first-level verification rules specify how to verify business indicators. Accordingly, the verification model can verify the corresponding business indicators according to the specific content in the first-level verification rules, thereby obtaining the verification result corresponding to the business indicator.

[0097] In some alternative implementations, the method further includes: Step c1: If the verification result indicates that the business indicator has failed the verification, then the verification rule base is searched based on the business indicator to determine the second-level verification rule corresponding to the business indicator. The verification granularity of the second-level verification rule is smaller than that of the first-level verification rule.

[0098] Step c2: Based on the second-level verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the new verification results corresponding to the business indicators.

[0099] Taking the business indicator of enterprise relationship under the verification dimension of authenticity as an example, its corresponding second-level verification rule can be: {Business indicator: Enterprise relationship judgment; Verification dimension: Authenticity Function: Verify whether two or more enterprises are related parties, and further judge when multiple enterprise names are inconsistent but similar; Input requirements: Two or more enterprise names; Execution conditions: Retrieve all enterprise names and output them; Use the enterprise relationship tool to query the relationship; If a relationship exists, the verification result is true; If no relationship exists, the verification result is false; Output format: Output in JSON format. Strictly follow the following JSON format to ensure that the key values ​​and order are consistent, and output the explanatory text at the same time: Enterprise relationship judgment: {Enterprise name: xxx (array showing the value in each contract); Verification result: true / false; Description: Explanatory text} Verification depth: Second level}.

[0100] If the verification result indicates that the business indicator has failed the verification, it means that the business indicator needs to be further reviewed. Since the verification granularity corresponding to the second-level verification rule is smaller than that corresponding to the first-level verification rule, the verification model is used to verify the business indicator according to the second-level verification rule in order to further explore the specific problems of the business indicator. In this way, more accurate new verification results can be obtained.

[0101] In some optional implementations, the method further includes: if the verification result indicates that the business indicator has passed the verification, then the verification process for the business indicator is terminated.

[0102] Step S604: Based on the verification results under each verification dimension, determine the background verification results corresponding to the business item.

[0103] Specifically, step S604 includes: Step S6041: If all verification results indicate that the business indicators corresponding to the verification results have passed the verification, then the background verification result indicating that the business matter has passed the verification is obtained.

[0104] As a specific example, taking the business indicator of the contract subject as an example, if the verification result is basically compliant, it means that the contract subject has passed the verification.

[0105] Step S6042: If any verification result indicates that the business indicator corresponding to the verification result has failed the verification, then the background verification result indicating that the business item has failed the verification is obtained.

[0106] As a specific example, taking the business indicator of contract subject as an example, if the verification result is "basically non-compliant", it means that the contract subject has not passed the verification.

[0107] As a concrete example, the importance of each verification dimension can be incorporated into the process of determining the background verification result for a business matter based on the verification results under various verification dimensions. For instance, the importance of the verification dimension of authenticity can be set higher than that of the verification dimensions of reasonableness and validity. If any verification result under the verification dimension of authenticity indicates that its corresponding business indicator has failed verification, then the background verification result is determined to be that the business matter has failed verification. If all verification results under the verification dimension of authenticity indicate that verification has passed, but any verification result under the verification dimensions of reasonableness or validity indicates that its corresponding business indicator has failed verification, then the background verification result is determined to be that the business matter has passed verification.

[0108] By adopting the logic that all business indicators must pass verification before a business item is deemed to have passed verification, it is possible to ensure that business items meet the requirements in all verification dimensions such as authenticity, validity, and reasonableness, thereby avoiding misjudgments caused by the omission of a single business indicator and improving the reliability of background verification results.

[0109] The business background verification method provided in this embodiment searches the verification rule base based on business indicators, which can quickly and accurately retrieve the first-level verification rules corresponding to the business indicators. Then, the verification model is used to perform preliminary verification of the business indicators according to the first-level verification rules. This can filter out obviously non-compliant business indicators in advance, reduce the amount of calculation for subsequent verification, and improve the efficiency and accuracy of the overall verification process.

[0110] As a specific application embodiment of the present invention, such as Figure 7 The diagram shows a specific business background verification method, which includes steps S701 to S708.

[0111] Step S701: Obtain the original background information corresponding to the business matter.

[0112] Step S702: Extract the business metrics corresponding to each verification dimension from the original background information.

[0113] Step S703: Search the verification rule base based on business indicators to determine the first-level verification rule corresponding to the business indicators.

[0114] Step S704: Verify the business indicators using the verification model according to the first-level verification rules to obtain the verification results corresponding to the business indicators.

[0115] Step S705: Determine whether the verification result indicates whether the business indicator corresponding to the verification result has passed the verification. If the verification result indicates that the business indicator corresponding to the verification result has failed the verification, proceed to steps S706 to S708; if the verification result indicates that the business indicator corresponding to the verification result has passed the verification, proceed to step S708.

[0116] Step S706: Search the verification rule base based on business indicators to determine the second-level verification rules corresponding to the business indicators.

[0117] Step S707: Use the verification model to verify the business indicators according to the second-level verification rules to obtain the new verification results corresponding to the business indicators.

[0118] Step S708: Based on the verification results or new verification results under each verification dimension, generate background verification results for the business matter.

[0119] This embodiment also provides a business background verification device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0120] This embodiment provides a business background verification device, such as Figure 8 As shown, it includes: The first acquisition module 810 is used to acquire the original background information corresponding to the business matter.

[0121] The first extraction module 820 is used to extract business indicators under each verification dimension from the original background information.

[0122] The first verification module 830 is used to verify business indicators based on the verification rules corresponding to the business indicators and to obtain the verification results corresponding to the business indicators.

[0123] The first determination module 840 is used to determine the background verification result corresponding to the business item based on the verification results under each verification dimension.

[0124] In some optional implementations, the first verification module 830 is further configured to retrieve the verification rule base based on the business indicators, determine the first-level verification rule corresponding to the business indicators, and store the mapping relationship between the business indicators and the verification rules; based on the first-level verification rule corresponding to the business indicators, the verification model is used to verify the business indicators to obtain the verification result corresponding to the business indicators.

[0125] In some alternative embodiments, the device further includes: The retrieval module is used to search the verification rule base based on the business indicator if the verification result indicates that the business indicator has failed the verification, and determine the second-level verification rule corresponding to the business indicator. The verification granularity of the second-level verification rule is smaller than that of the first-level verification rule.

[0126] The second verification module is used to verify the business indicators based on the second-level verification rules corresponding to the business indicators, and to obtain the new verification results corresponding to the business indicators.

[0127] In some optional implementations, the first extraction module 820 is further configured to obtain target prompt words corresponding to each extraction model; for each extraction model, the target prompt words are used to guide the extraction model to extract initial business indicators under each verification dimension in the original background information; for each verification dimension, the initial business indicators extracted by all extraction models corresponding to the verification dimension are compared to obtain the business indicators under the verification dimension, thereby obtaining the business indicators corresponding to each verification dimension.

[0128] In some optional implementations, the first extraction module 820 is further configured to compare the first initial business indicator corresponding to the first extraction model with the second initial business indicator corresponding to the second extraction model to obtain a comparison result; if the comparison result indicates that the first initial business indicator and the second initial business indicator are the same, then the first initial business indicator or the second initial business indicator is determined as the business indicator under the verification dimension; if the comparison result indicates that the first initial business indicator and the second initial business indicator are different, then a new first initial business indicator is re-extracted using the first extraction model, and the business indicator under the verification dimension is re-determined based on the new first initial business indicator and the second initial business indicator.

[0129] In some optional implementations, the device for acquiring the target prompt word corresponding to each extraction model includes: a second acquisition module, used to acquire multiple training background information, the label business indicators corresponding to each training background information, and the initial prompt word corresponding to the extraction model; a second extraction module, used to guide the extraction model to extract the training business indicators from each training background information using the initial prompt word; a comparison module, used to compare the label business indicators and training business indicators corresponding to each training background information to obtain the accuracy corresponding to each training background information; an adjustment module, used to adjust the initial prompt word to obtain the target prompt word if there is an accuracy less than a preset threshold; and a second determination module, used to determine the initial prompt word as the target prompt word if all accuracy rates are greater than or equal to the preset threshold.

[0130] In some optional implementations, the first determining module 840 is further configured to obtain a background verification result indicating that the business item has passed verification if all verification results indicate that the corresponding business indicator has passed verification; and to obtain a background verification result indicating that the business item has failed verification if any verification result indicates that the corresponding business indicator has failed verification.

[0131] The business background verification device provided in this embodiment of the invention can execute the business background verification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0132] Figure 9 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. The electronic device may be the terminal device or server described above.

[0133] The following is a detailed reference. Figure 9 This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 901, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 902 or a program loaded from memory 908 into random access memory (RAM) 903. The RAM 903 also stores various programs and data required for the operation of the electronic device. The processor 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0134] Typically, the following devices can be connected to I / O interface 905: input devices 906 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 907 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 908 including, for example, magnetic tapes, hard disks, etc.; and communication devices 909. Communication device 909 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 9 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0135] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 909, or installed from a memory 908, or installed from a ROM 902. When the computer program is executed by the processor 901, it performs the functions defined in the business background verification method of the embodiments of the present invention.

[0136] Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0137] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the business background verification method shown in the above embodiments is implemented.

[0138] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0139] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A business background verification method, characterized in that, The method includes: Obtain the original background information corresponding to the business transaction; Extract business metrics under each verification dimension from the original background information; Based on the verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the verification results corresponding to the business indicators. Based on the verification results under each of the verification dimensions, the background verification result corresponding to the business matter is determined.

2. The method according to claim 1, characterized in that, The process of verifying the business indicators based on the verification rules corresponding to the business indicators, and obtaining the verification results corresponding to the business indicators, includes: Based on the business indicators, the verification rule base is retrieved to determine the first-level verification rule corresponding to the business indicators. The verification rule base is used to store the mapping relationship between the business indicators and the first-level verification rules. Based on the first-level verification rules corresponding to the business indicators, the verification model is used to verify the business indicators and obtain the verification results corresponding to the business indicators.

3. The method according to claim 2, characterized in that, The method further includes: If the verification result indicates that the business indicator has failed the verification, then the verification rule base is searched based on the business indicator to determine the second-level verification rule corresponding to the business indicator. The verification granularity corresponding to the second-level verification rule is smaller than the verification granularity corresponding to the first-level verification rule. Based on the second-level verification rules corresponding to the business indicators, the verification model is used to verify the business indicators to obtain new verification results corresponding to the business indicators.

4. The method according to claim 1, characterized in that, The step of extracting business metrics under each verification dimension from the original background information includes: Obtain the target prompt words corresponding to each extraction model; For each extraction model, the target prompt words are used to guide the extraction model to extract the initial business indicators under each verification dimension in the original background information; For each verification dimension, the initial business indicators extracted by all the extraction models corresponding to the verification dimension are compared to obtain the business indicators under the verification dimension, thereby obtaining the business indicators corresponding to each verification dimension.

5. The method according to claim 4, characterized in that, Each extraction model includes a first extraction model and a second extraction model; the step of comparing the initial business indicators extracted by all extraction models corresponding to the verification dimension to obtain the business indicators under the verification dimension includes: The first initial business indicator corresponding to the first extraction model is compared with the second initial business indicator corresponding to the second extraction model to obtain the comparison result; If the comparison result indicates that the first initial business indicator and the second initial business indicator are the same, then the first initial business indicator or the second initial business indicator is determined as the business indicator under the verification dimension. If the comparison result indicates that the first initial business indicator and the second initial business indicator are different, then the first extraction model is used to re-extract the new first initial business indicator, and the business indicator under the verification dimension is determined based on the new first initial business indicator and the second initial business indicator.

6. The method according to claim 4, characterized in that, The process of obtaining the target prompt word corresponding to each extraction model includes: Acquire multiple training background information, the label business indicators corresponding to each training background information, and the initial prompt words corresponding to the extraction model; The initial prompt words are used to guide the extraction model to extract training business metrics from each of the training background information. The accuracy corresponding to each training background information is obtained by comparing the label business indicator and the training business indicator. If there is an accuracy rate lower than a preset threshold, the initial prompt word is adjusted to obtain the target prompt word; If all the accuracy rates are greater than or equal to the preset threshold, then the initial prompt word is determined as the target prompt word.

7. The method according to any one of claims 1 to 6, characterized in that, The determination of the background verification result corresponding to the business matter based on the verification results under each of the verification dimensions includes: If all the verification results indicate that the business indicators corresponding to the verification results have passed the verification, then the background verification result indicating that the business matter has passed the verification is obtained; If any of the verification results indicates that the business indicator corresponding to the verification result has failed the verification, then the background verification result indicating that the business matter has failed the verification is obtained.

8. A business background verification device, characterized in that, The device includes: The first acquisition module is used to acquire the original background information corresponding to the business item; The first extraction module is used to extract business indicators under each verification dimension from the original background information; The first verification module is used to verify the business indicators based on the verification rules corresponding to the business indicators and using the verification model to obtain the verification results corresponding to the business indicators. The first determining module is used to determine the background verification result corresponding to the business matter based on the verification results under each of the verification dimensions.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the business background verification method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the business background verification method according to any one of claims 1 to 7.

Images