Credit certificate examination method, device, equipment and medium
By extracting the elements to be reviewed in the letter of credit review method and performing fine-grained splits, the problem that the elements to be reviewed in the existing technology cannot meet the needs of letter of credit review, and the accuracy of letter of credit review is improved.
Patent Information
- Application Number
- CN202411665868.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-20
- Publication Date
- 2025-03-11
AI Technical Summary
In the existing methods of letter of credit review, the elements to be reviewed cannot meet the needs of letter of credit review, resulting in low accuracy of letter of credit review.
By receiving an audit request carrying the audit logo and the file to be reviewed, the elements to be reviewed are extracted from the file to be reviewed, and fine-grained splits are performed based on the feature type to which each composite element belongs to, to obtain fine-grained elements. Then, the audit results of the documents to be reviewed are determined based on the monolithic elements and fine-grained elements.
Through fine-grained splitting and standardized processing, the fine-grained needs of letters of credit review are met, and the accuracy of letters of credit review is improved.
Smart Images
Figure CN119599599A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing, and in particular to a method, device, equipment and medium for examining a letter of credit. Background Art
[0002] Document review under a letter of credit is a comprehensive service provided by banks to exporters under a letter of credit, which integrates document review and document collection. When reviewing a letter of credit, it is necessary to follow international practices and practical rules and follow the principles of "document consistency", "document consistency" and "consistency with rules".
[0003] In the specific implementation, it is necessary to use a predefined label system to mark the letters of credit and documents, and then extract the elements to be reviewed of the letters of credit and documents through the natural language processing model, and then review the elements to be reviewed according to the preset document review rules.
[0004] However, the existing labeling system mainly focuses on how to reduce the difficulty of labeling, which results in the extracted elements to be reviewed being unable to meet the requirements of letter of credit review, affecting the accuracy of letter of credit review. Summary of the invention
[0005] The embodiments of the present application provide a method, device, equipment and medium for reviewing letters of credit, which are used to solve the problem that the elements to be reviewed cannot meet the requirements of letter of credit review, resulting in low accuracy of letter of credit review.
[0006] In a first aspect, an embodiment of the present application provides a method for examining a letter of credit, the method comprising:
[0007] When receiving an audit request carrying an audit mark and a document to be audited, extracting the elements to be audited from the document to be audited, wherein the audit mark is used to indicate that the audit request is a single audit request or a document audit request, the single audit request is used to audit the document and the document, the document audit request is used to audit the document and the letter of credit, and the elements to be audited include single elements and composite elements;
[0008] Based on the element type to which each composite element belongs, fine-grained splitting of the composite elements in the element to be reviewed is performed to obtain fine-grained elements;
[0009] The audit result of the document to be audited is determined according to the monomer elements and the fine-grained elements.
[0010] In an embodiment of the present application, when an audit request carrying an audit mark and a document to be audited is received, the elements to be audited are extracted from the document to be audited, and based on the element type to which each composite element in the elements to be audited belongs, the composite elements are finely decomposed to obtain fine-grained elements, thereby meeting the fine-grained requirements during the review of letters of credit. The audit results of the document to be audited are determined based on the individual elements and the fine-grained elements, which is conducive to improving the accuracy of the review of letters of credit.
[0011] In a possible embodiment, the fine-grained splitting of the composite elements in the elements to be reviewed based on the element type to which each composite element belongs to obtains fine-grained elements, including:
[0012] For any composite element in the elements to be reviewed, a public parsing method matching the element type to which the composite element belongs is called to perform fine-grained decomposition on the composite element to obtain fine-grained elements.
[0013] In a possible embodiment, before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes:
[0014] For any fine-grained element obtained by splitting, the fine-grained element is standardized according to the element type to which the composite element corresponding to the fine-grained element belongs;
[0015] The standardization process includes part or all of text cleaning, date format conversion, numerical format conversion and Boolean value type conversion.
[0016] In a possible embodiment, before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes:
[0017] When the audit identifier indicates that the audit request is a document audit request, determining candidate elements to be audited from the monomer elements and the fine-grained elements, wherein the candidate elements to be audited are elements to be audited extracted from the letter of credit;
[0018] Using the pre-stored entity name, the common elements in the candidate elements to be reviewed are replaced by references, and the common elements are the reference information used for general description in the terms and conditions of the letter of credit.
[0019] In a possible embodiment, before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes:
[0020] According to the pre-stored entity name, a link identifier is added to the entity element in the monomer element and the fine-grained element, and the link identifier is used to indicate that the entity element is a known entity element.
[0021] In a possible embodiment, determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element includes:
[0022] According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements;
[0023] The audit elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements to obtain the audit result of the file to be audited.
[0024] In a possible embodiment, determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element includes:
[0025] According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements;
[0026] When the audit identifier indicates that the audit request is a document audit request, candidate audit elements are determined from the audit elements, where the candidate audit elements are obtained by structurally processing the elements to be audited extracted from the document;
[0027] For at least one candidate audit element configured with an inference identifier, determining a target element not included in the candidate audit element according to the inference identifier corresponding to each candidate audit element, the at least one candidate audit element belonging to the same element type, the inference identifier being used to indicate the priority of each candidate audit element in the process of determining the target element;
[0028] The audit elements and the target elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements and the target elements to obtain an audit result of the file to be audited.
[0029] In a possible embodiment, determining the target element not included in the candidate audit element according to the inferred identifier corresponding to each candidate audit element includes:
[0030] Determine the target inference identifier with the highest priority among the inference identifiers corresponding to each candidate audit element;
[0031] The element value of the candidate audit element corresponding to the target inference identifier is used as the element value of the target element.
[0032] In a second aspect, the present application provides a letter of credit document review device, the device comprising:
[0033] An element extraction module is used to extract the elements to be reviewed from the files to be reviewed when receiving an audit request carrying an audit mark and a file to be reviewed, wherein the audit mark is used to indicate that the audit request is a single audit request or a document audit request, the single audit request is used to audit the documents and the documents, the document audit request is used to audit the documents and the letter of credit, and the elements to be reviewed include single elements and composite elements;
[0034] A fine-grained splitting module, used for performing fine-grained splitting of the composite elements in the elements to be reviewed based on the element type to which each composite element belongs, to obtain fine-grained elements;
[0035] The audit module is used to determine the audit result of the file to be audited according to the monomer elements and the fine-grained elements.
[0036] In a possible embodiment, the fine-grained splitting module is specifically used to:
[0037] For any composite element in the elements to be reviewed, a public parsing method matching the element type to which the composite element belongs is called to perform fine-grained decomposition on the composite element to obtain fine-grained elements.
[0038] In a possible embodiment, the audit module is further used to:
[0039] For any fine-grained element obtained by splitting, the fine-grained element is standardized according to the element type to which the composite element corresponding to the fine-grained element belongs;
[0040] The standardization process includes part or all of text cleaning, date format conversion, numerical format conversion and Boolean value type conversion.
[0041] In a possible embodiment, the audit module is further used to:
[0042] When the audit identifier indicates that the audit request is a document audit request, determining candidate elements to be audited from the monomer elements and the fine-grained elements, wherein the candidate elements to be audited are elements to be audited extracted from the letter of credit;
[0043] Using the pre-stored entity name, the common elements in the candidate elements to be reviewed are replaced by references, and the common elements are the reference information used for general description in the terms and conditions of the letter of credit.
[0044] In a possible embodiment, the audit module is further used to:
[0045] According to the pre-stored entity name, a link identifier is added to the entity element in the monomer element and the fine-grained element, and the link identifier is used to indicate that the entity element is a known entity element.
[0046] In a possible embodiment, the audit module is specifically used to:
[0047] According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements;
[0048] The audit elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements to obtain the audit result of the file to be audited.
[0049] In a possible embodiment, the audit module is specifically used to:
[0050] According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements;
[0051] When the audit identifier indicates that the audit request is a document audit request, candidate audit elements are determined from the audit elements, where the candidate audit elements are obtained by structurally processing the elements to be audited extracted from the document;
[0052] For at least one candidate audit element configured with an inference identifier, determining a target element not included in the candidate audit element according to the inference identifier corresponding to each candidate audit element, the at least one candidate audit element belonging to the same element type, the inference identifier being used to indicate the priority of each candidate audit element in the process of determining the target element;
[0053] The audit elements and the target elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements and the target elements to obtain an audit result of the file to be audited.
[0054] In a possible embodiment, the audit module is specifically used to:
[0055] The step of determining, based on the inferred identifier corresponding to each candidate audit element, a target element not included in the candidate audit element, includes:
[0056] Determine the target inference identifier with the highest priority among the inference identifiers corresponding to each candidate audit element;
[0057] The element value of the candidate audit element corresponding to the target inference identifier is used as the element value of the target element.
[0058] In a third aspect, the present application provides an electronic device, including:
[0059] A memory for storing program instructions;
[0060] A processor is used to call the program instructions stored in the memory, and execute the steps included in the letter of credit examination method described in any one of the first aspects according to the obtained program instructions.
[0061] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the letter of credit review method described in any one of the first aspects.
[0062] In a fifth aspect, the present application provides a computer program product, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the letter of credit examination method described in any one of the first aspects.
[0063] The technical effects brought about by any one of the implementation methods of the second to fifth aspects can refer to the technical effects brought about by the implementation method of the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] Figure 1 A schematic diagram of an application scenario of a method for examining a letter of credit provided in an embodiment of the present application;
[0065] Figure 2 A flowchart of a method for examining a letter of credit provided in an embodiment of the present application;
[0066] Figure 3 A schematic diagram of element types in an invoice document provided in an embodiment of the present application;
[0067] Figure 4 A schematic diagram of a mapping configuration table provided in an embodiment of the present application;
[0068] Figure 5 A structural diagram of a letter of credit review device provided in an embodiment of the present application;
[0069] Figure 6 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0070] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. Obviously, the described embodiment is only a part of the embodiment of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be arbitrarily combined with each other. In addition, although the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in an order different from that here.
[0071] The terms "first" and "second" in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the term "comprising" and any of their variations are intended to cover non-exclusive protection. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices. "Multiple" in the present application can mean at least two, for example, two, three or more, and the embodiments of the present application are not limited.
[0072] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description. It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.
[0073] In the technical solution of this application, the acquisition, transmission, storage, and use of data are in compliance with the requirements of relevant national laws and regulations.
[0074] Before introducing the letter of credit review method provided by the embodiment of the present application, in order to facilitate understanding, the technical background of the embodiment of the present application is first introduced in detail below.
[0075] Document review under a letter of credit is a comprehensive service provided by banks to exporters under a letter of credit, which integrates document review and document collection. When reviewing a letter of credit, it is necessary to follow international practices and practical rules and follow the principles of "document consistency", "document consistency" and "consistency with rules".
[0076] In the specific implementation, it is necessary to use a predefined labeling system to label letters of credit and documents, and then extract the elements to be reviewed of the letters of credit and documents through the natural language processing model, and then review the elements to be reviewed according to the preset document review rules. In related technologies, in order to reduce the difficulty of labeling and the difficulty of training natural language processing models, it is usually necessary to adopt a concise and high-labeling labeling system.
[0077] However, the elements to be reviewed extracted by the natural language processing model trained based on the traditional label system cannot meet the needs of single document review and document review, which affects the accuracy of letter of credit review.
[0078] In view of this, the embodiments of the present application provide a method, device, equipment and medium for reviewing letters of credit, which are used to solve the problem that the elements to be reviewed cannot meet the requirements of letter of credit review, resulting in low accuracy of letter of credit review.
[0079] The inventive concept of the present application can be summarized as follows: when an audit request carrying an audit mark and a document to be audited is received, the elements to be audited are extracted from the document to be audited, and based on the element type to which each composite element in the elements to be audited belongs, the composite elements are finely split into fine-grained elements to obtain fine-grained elements, thereby meeting the fine-grained requirements during the review of letters of credit, and the audit results of the document to be audited are determined based on the individual elements and the fine-grained elements, which is conducive to improving the accuracy of the review of letters of credit.
[0080] After introducing the inventive concept of this application, Figure 1 , the application scenarios of the embodiments of the present application are introduced.
[0081] Reference Figure 1 , is an application scenario diagram of a method for examining a letter of credit provided in an embodiment of the present application. Figure 1 As shown, the scenario includes: a terminal device 110 and a server 120, and the terminal device 110 and the server 120 can communicate with each other through a communication network.
[0082] In an optional implementation, the communication network may be a wired network or a wireless network. Therefore, the terminal device 110 and the server 120 may be directly or indirectly connected via a wired or wireless communication. For example, the terminal device 110 may be indirectly connected to the server 120 via a wireless access point, or the terminal device 110 may be directly connected to the server 120 via the Internet, and this application does not limit this.
[0083] In the embodiment of the present application, the terminal device 110 includes but is not limited to mobile phones, tablet computers, laptop computers, desktop computers, e-book readers, intelligent voice interaction devices, smart home appliances, car terminals and other devices.
[0084] The server 120 is a background server corresponding to the terminal device 110. The server 120 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.
[0085] It should be noted that the letter of credit review method in the embodiment of the present application can be deployed in an electronic device, which can be a server or a terminal device.
[0086] The server can be Figure 1 The server 120 shown in FIG. Figure 1 The terminal device may be other servers other than the server 120 shown in FIG. Figure 1 The terminal device 110 shown in FIG. 1 may also be Figure 1 Other terminal devices other than the terminal device 110 shown in FIG.
[0087] Figure 1 The figure is only an example. In fact, the number of terminal devices 110 and servers 120 is not limited and is not specifically limited in the embodiments of the present application.
[0088] Based on the above application scenarios, the following describes the letter of credit review method provided by the exemplary embodiment of the present application in combination with the above-described application scenarios according to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation methods of the present application are not limited in this regard.
[0089] refer to Figure 2 , Figure 2 An exemplary method for examining a letter of credit in an embodiment of the present application is provided. The method can be executed by an electronic device and includes the following steps:
[0090] Step S201: when receiving an audit request carrying an audit mark and a file to be audited, extracting elements to be audited from the file to be audited.
[0091] Among them, the audit mark is used to indicate that the audit request is a single audit request or a document audit request. The single audit request is used to audit the document and the document, and the document audit request is used to audit the document and the letter of credit.
[0092] In a specific implementation, when the audit mark indicates that the audit request is a single audit request, the document and the elements of the document to be audited can be extracted through natural language processing technology. When the audit mark indicates that the audit request is a document audit request, the elements of the document and letter of credit to be audited can be extracted through natural language processing technology.
[0093] Among them, the elements to be reviewed include single elements and composite elements. For example, the "invoiceDate": "XXXXXXXX" (invoice issuance date) in the invoice document is a single element that cannot be fine-grained. However, "GOODSAMT": "USD12,000.00" (goods amount) is a composite element that can be fine-grained, which can be split into "GOODSAMT.currency.name": "USD" (the currency name of the goods amount) and "GOODSAMT.value": "12,000.00" (the value of the goods amount).
[0094] It should be noted that the elements to be reviewed usually include single elements and composite elements, but in some embodiments, the elements to be reviewed may only include composite elements, or may only include single elements.
[0095] Step S202 , based on the element type to which each composite element belongs, fine-grained decomposition is performed on the composite elements in the elements to be reviewed to obtain fine-grained elements.
[0096] In response to the audit requirements of single and document audits, this application defines the element types corresponding to single elements and fine-grained elements to complete the subsequent processing steps. Taking invoice documents as an example, some element types can be found in Figure 3 shown.
[0097] Furthermore, based on the element type to which each composite element belongs, when fine-grained splitting of the composite elements in the audited elements is performed, Figure 3 The abstract feature type in the model is used to perform fine-grained decomposition of composite features.
[0098] In a specific implementation, for any composite element in the elements to be reviewed, a public parsing method that matches the element type to which the composite element belongs can be called to perform fine-grained decomposition on the composite element to obtain fine-grained elements.
[0099] For example, let's assume that a certain composite element extracted is "PRICE_TERM": "CIF BUSAN PORT" (transaction terms: the seller bears the cost of delivering the goods to Busan Port). Then, the public parsing method fun_priceterm() corresponding to PriceTerm can be called to perform fine-grained splitting of the composite element. The result of the splitting is "PRICE_TERM.tradeTerm.name": "CIF" (transaction terms name: the seller bears the cost), "PRICE_TERM.pricePlace.name": "BUSAN PORT (transaction terms place name: Busan Port).
[0100] In an optional implementation, after the composite elements in the document to be reviewed are split into fine-grained elements, any fine-grained element obtained by the splitting can be standardized according to the element type to which the composite element corresponding to the fine-grained element belongs.
[0101] In some embodiments, if the fine-grained element belongs to the Party entity type, the fine-grained element may be subjected to text cleaning, and the entity of the Party entity type may be cleaned of irrelevant words before and after.
[0102] In other embodiments, if the fine-grained element is of the Amount type, since the elements of the Amount type need to be reviewed for amount consistency and amount value, the numerical format of the Amount type can be converted to a standard two-digit decimal place.
[0103] In some other embodiments, if the fine-grained element is of the Date type, the date format of the fine-grained element may be converted into a standard 8-bit date.
[0104] In other embodiments, the fine-grained elements may be converted to Boolean type, integer type, etc. according to the element type to which the composite element to which the fine-grained element corresponds belongs.
[0105] It should be noted that this application does not impose any restrictions on the specific implementation method of standardized processing. Since different documents issued by different companies or the same company may appear in multiple formats, standardizing the fine-grained elements obtained by splitting can improve the efficiency and accuracy of subsequent letter of credit review.
[0106] In an optional implementation, for the elements that need to be replaced by reference, the processed entity name can be pre-stored in the electronic device, and the entity name can be used to replace the element value.
[0107] In specific implementation, in the description of the terms of a letter of credit, some common elements are often used as reference information for general description.
[0108] For example, in the terms of the letter of credit, the "Applicant" in "notify Applicant" is a general element, while the specific company name is under the "notify party" column in the document. At this time, the "Applicant" in the terms of the letter of credit can be replaced with the company's entity name, so as to compare it with the company name in the document for consistency.
[0109] In some embodiments, a preset mapping configuration table may be used to record the processing flow of whether the element to be reviewed needs to be replaced by a reference.
[0110] For example, the mapping configuration table may include data sources, single elements / fine-grained elements, audit elements, element types, and various types of operations to be processed. Part of the content of the mapping configuration table can be found in Figure 4 shown.
[0111] like Figure 4 As shown, "APPLICANT" in "NOTIFYPARTY.referenceIndicator": "APPLICANT" needs to be replaced by a reference. The specific applicant information can be added to notifyParty as an element value. Assuming that the processed entity name is ABC CO LTD, after the reference replacement, we can get "NOTIFYPARTY.referenceIndicator": "ABC CO LTD".
[0112] In an optional implementation, the entity name pre-stored in the database may be used to identify the entity. Figure 4 For each element that needs to be entity linked and is listed in the mapping configuration table, a link identifier is added, and the link identifier is used to indicate that the entity element is a known entity element.
[0113] For example, the same company entity or the same port entity may have different forms of field representation. Entity linking processing for known entity names can improve the accuracy of letter of credit review and avoid misjudgment in subsequent content consistency reviews of company entity types and inconsistency reviews of port entity types.
[0114] Step S203, determining the audit result of the document to be audited according to the monomer elements and the fine-grained elements.
[0115] In an optional embodiment, according to Figure 4The mapping configuration table shown performs structured processing on monomer elements and fine-grained elements to obtain audit elements, and then sends the audit elements to the audit engine, so that the audit engine performs audits based on the audit elements and preset audit rules to obtain the audit results of the files to be audited.
[0116] In another optional implementation, when the review identifier indicates that the review request is a document review request, the description of the terms of the letter of credit may contain target elements that are not included in the document.
[0117] At this time, according to Figure 4 The mapping configuration table shown performs structured processing on monomer elements and fine-grained elements. After obtaining the audit elements, the target elements not included in the candidate audit elements can be determined based on the inferred identifier corresponding to each candidate audit element, where the candidate audit elements are obtained by structured processing of the elements to be audited extracted from the document.
[0118] In a specific implementation, the target inference identifier with the highest priority can be determined among the inference identifiers corresponding to each candidate audit element, and then the element value of the candidate audit element corresponding to the target inference identifier is used as the element value of the target element to obtain the inferred target element.
[0119] For example, assuming that the invoice contains the element "letterHead.name": "DEF CO LTD" (the name of the letterhead is: DEF CO LTD), and the element "beneficiary.name": "HIJ CO LTD" (the name of the beneficiary is: HIJ COLTD). Further, it can be determined that the inferred identifier of the element "letterHead.name": "DEF CO LTD" is "ben-1" and the inferred identifier of the element "beneficiary.name": "HIJ CO LTD" is "ben-4". It can be determined that the priority of the target element "real beneficiary" in the process letterHeader is greater than beneficiary, that is, "ben-1" is the target inferred identifier, and the element value of the audit element corresponding to the target inferred identifier can be used as the element value of the target element, that is, the target element "ben": "DEF CO LTD" is obtained.
[0120] After obtaining the inferred target elements through the target inference identifier, the audit elements and the target elements can be sent to the audit engine together, so that the audit engine performs an audit based on the audit elements, target elements and preset audit rules, so as to obtain more accurate audit results.
[0121] Based on the foregoing description, in an embodiment of the present application, when an audit request carrying an audit mark and a document to be audited is received, the elements to be audited are extracted from the document to be audited, and based on the element type to which each composite element in the elements to be audited belongs, the composite elements are finely decomposed to obtain fine-grained elements, thereby meeting the fine-grained requirements during the review of letters of credit. The audit results of the document to be audited are determined based on the individual elements and the fine-grained elements, which is conducive to improving the accuracy of the review of letters of credit.
[0122] Moreover, without changing the traditional labeling system, the present application ensures the recognition effect of the natural language processing model. Through a preset mapping configuration table, each element mark needs to be configured, mapped, and processed. By standardizing, referencing, entity linking, structured mapping, and inferring the target elements, the accuracy of letter of credit review can be further improved.
[0123] Based on the same inventive concept, the present application embodiment provides a letter of credit review device, please refer to Figure 5 The device comprises: an element extraction module 501, a fine-grained splitting module 502 and an auditing module 503, wherein:
[0124] The element extraction module 501 is used to extract the elements to be reviewed from the files to be reviewed when receiving an audit request carrying an audit mark and files to be reviewed, wherein the audit mark is used to indicate that the audit request is a single audit request or a document audit request, the single audit request is used to audit the documents and the documents, the document audit request is used to audit the documents and the letter of credit, and the elements to be reviewed include single elements and composite elements.
[0125] The fine-grained splitting module 502 is used to perform fine-grained splitting on the composite elements in the elements to be reviewed based on the element type to which each composite element belongs, so as to obtain fine-grained elements.
[0126] The audit module 503 is used to determine the audit result of the file to be audited according to the monomer elements and the fine-grained elements.
[0127] In a possible embodiment, the fine-grained splitting module 502 is specifically used to:
[0128] For any composite element in the elements to be reviewed, a public parsing method matching the element type to which the composite element belongs is called to perform fine-grained decomposition on the composite element to obtain fine-grained elements.
[0129] In a possible embodiment, the audit module 503 is further configured to:
[0130] For any fine-grained element obtained by splitting, the fine-grained element is standardized according to the element type to which the composite element corresponding to the fine-grained element belongs;
[0131] Among them, standardization processing includes part or all of text cleaning, date format conversion, numerical format conversion and Boolean type conversion.
[0132] In a possible embodiment, the audit module 503 is further configured to:
[0133] When the audit flag indicates that the audit request is a document audit request, candidate elements to be audited are determined in the monomer elements and fine-grained elements, and the candidate elements to be audited are elements to be audited extracted from the letter of credit;
[0134] Use the pre-stored entity names to replace the common elements in the candidate elements to be reviewed. The common elements are the reference information used for general descriptions in the terms and conditions of the letter of credit.
[0135] In a possible embodiment, the audit module 503 is further configured to:
[0136] According to the pre-stored entity name, a link identifier is added to the entity element in the monomer element and the fine-grained element, and the link identifier is used to indicate that the entity element is a known entity element.
[0137] In a possible embodiment, the audit module 503 is specifically used to:
[0138] According to the preset mapping relationship, the monomer elements and fine-grained elements are structured to obtain the audit elements;
[0139] The audit elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements and obtains the audit results of the files to be audited.
[0140] In a possible embodiment, the audit module 503 is specifically configured to:
[0141] According to the preset mapping relationship, the monomer elements and fine-grained elements are structured to obtain the audit elements;
[0142] When the audit indicator indicates that the audit request is a document audit request, candidate audit elements are determined in the audit elements. The candidate audit elements are obtained by structurally processing the elements to be audited extracted from the document.
[0143] For at least one candidate audit element configured with an inference identifier, determining a target element not included in the candidate audit element according to the inference identifier corresponding to each candidate audit element, at least one candidate audit element belongs to the same element type, and the inference identifier is used to indicate the priority of each candidate audit element in the process of determining the target element;
[0144] The audit elements and target elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements and target elements to obtain the audit results of the file to be audited.
[0145] In a possible embodiment, the audit module 503 is specifically configured to:
[0146] According to the inferred identifier corresponding to each candidate audit element, determine the target elements not included in the candidate audit elements, including:
[0147] Among the inference identifiers corresponding to each candidate audit element, determine the target inference identifier with the highest priority;
[0148] The element value of the candidate audit element corresponding to the target inference identifier is used as the element value of the target element.
[0149] The letter of credit review device provided in the embodiment of the present application and the letter of credit review method in the above embodiment have the same beneficial effects, which will not be described one by one here.
[0150] After introducing the letter of credit examination method and letter of credit examination device according to the exemplary embodiments of the present application, next, an electronic device provided according to an embodiment of the present application is introduced.
[0151] The present application embodiment provides an electronic device, which can implement the credit document examination method discussed above. Figure 6 , the device includes a memory 601, a processor 602 and a bus 603.
[0152] The memory 601 is used to store computer programs executed by the processor 602. The memory 601 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and programs required for running the instant messaging function, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.
[0153] The memory 601 may be a volatile memory, such as a random-access memory (RAM); the memory 601 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), or the memory 601 may be any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 601 may be a combination of the above memories.
[0154] The processor 602 may include one or more central processing units (CPU) or a digital processing unit, etc. The processor 602 is used to implement the letter of credit examination method in the above embodiment when calling the computer program stored in the memory 601.
[0155] The specific connection medium between the memory 601 and the processor 602 is not limited in the embodiment of the present application. Figure 6 In the embodiment, the memory 601 and the processor 602 are connected via a bus 603. The bus 603 is connected to the processor 602 via a bus 603. Figure 6 The connections between other components are shown in bold lines, which are only for illustration and are not intended to be limiting. Bus 603 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0156] Based on the same inventive concept, the embodiment of the present application provides a computer-readable storage medium, and the computer program product includes: computer program code, when the computer program code is run on a computer, the computer executes any of the letter of credit examination methods discussed above. Since the principle of solving the problem by the above-mentioned computer-readable storage medium is similar to that of the letter of credit examination method, the implementation of the above-mentioned computer-readable storage medium can refer to the implementation of the method, and the repeated parts will not be repeated.
[0157] Based on the same inventive concept, the embodiment of the present application also provides a computer program product, which includes: computer program code, when the computer program code is run on a computer, the computer executes any of the letter of credit examination methods discussed above. Since the principle of solving the problem by the above computer program product is similar to that of the letter of credit examination method, the implementation of the above computer program product can refer to the implementation of the method, and the repeated parts will not be repeated.
[0158] It should be understood by those skilled in the art that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0159] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0160] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0161] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of user-operated steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0162] Obviously, a person skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A method for examining a letter of credit, characterized in that: The method comprises: When receiving an audit request carrying an audit mark and a document to be audited, extracting the elements to be audited from the document to be audited, wherein the audit mark is used to indicate that the audit request is a single audit request or a document audit request, the single audit request is used to audit the document and the document, the document audit request is used to audit the document and the letter of credit, and the elements to be audited include single elements and composite elements; Based on the element type to which each composite element belongs, fine-grained splitting of the composite elements in the element to be reviewed is performed to obtain fine-grained elements; The audit result of the document to be audited is determined according to the monomer elements and the fine-grained elements.
2. The method according to claim 1, characterized in that Based on the element type to which each composite element belongs, the composite elements in the element to be reviewed are finely divided to obtain fine-grained elements, including: For any composite element in the elements to be reviewed, a public parsing method matching the element type to which the composite element belongs is called to perform fine-grained decomposition on the composite element to obtain fine-grained elements.
3. The method according to claim 1, characterized in that Before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes: For any fine-grained element obtained by splitting, the fine-grained element is standardized according to the element type to which the composite element corresponding to the fine-grained element belongs; The standardization process includes part or all of text cleaning, date format conversion, numerical format conversion and Boolean value type conversion.
4. The method according to claim 1, characterized in that: Before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes: When the audit identifier indicates that the audit request is a document audit request, determining candidate elements to be audited from the monomer elements and the fine-grained elements, wherein the candidate elements to be audited are elements to be audited extracted from the letter of credit; Using the pre-stored entity name, the common elements in the candidate elements to be reviewed are replaced by references, and the common elements are the reference information used for general description in the terms and conditions of the letter of credit.
5. The method according to claim 1, characterized in that Before determining the audit result of the to-be-audited file according to the monomer element and the fine-grained element, the method further includes: According to the pre-stored entity name, a link identifier is added to the entity element in the monomer element and the fine-grained element, and the link identifier is used to indicate that the entity element is a known entity element.
6. The method according to claim 1, characterized in that Determining the audit result of the to-be-audited document according to the monomer element and the fine-grained element includes: According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements; The audit elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements to obtain the audit result of the file to be audited.
7. The method according to claim 1, characterized in that Determining the audit result of the to-be-audited document according to the monomer element and the fine-grained element includes: According to the preset mapping relationship, the monomer elements and the fine-grained elements are structured to obtain audit elements; When the audit identifier indicates that the audit request is a document audit request, candidate audit elements are determined from the audit elements, where the candidate audit elements are obtained by structurally processing the elements to be audited extracted from the document; For at least one candidate audit element configured with an inference identifier, determining a target element not included in the candidate audit element according to the inference identifier corresponding to each candidate audit element, the at least one candidate audit element belonging to the same element type, the inference identifier being used to indicate the priority of each candidate audit element in the process of determining the target element; The audit elements and the target elements are sent to the audit engine, so that the audit engine performs an audit based on the audit elements and the target elements to obtain an audit result of the file to be audited.
8. The method according to claim 7, characterized in that The step of determining, based on the inferred identifier corresponding to each candidate audit element, a target element not included in the candidate audit element, includes: Determine the target inference identifier with the highest priority among the inference identifiers corresponding to each candidate audit element; The element value of the candidate audit element corresponding to the target inference identifier is used as the element value of the target element.
9. A letter of credit examination device, characterized in that: The device comprises: An element extraction module is used to extract the elements to be reviewed from the files to be reviewed when receiving an audit request carrying an audit mark and a file to be reviewed, wherein the audit mark is used to indicate that the audit request is a single audit request or a document audit request, the single audit request is used to audit the documents and the documents, the document audit request is used to audit the documents and the letter of credit, and the elements to be reviewed include single elements and composite elements; A fine-grained splitting module, used for performing fine-grained splitting of the composite elements in the elements to be reviewed based on the element type to which each composite element belongs, to obtain fine-grained elements; The audit module is used to determine the audit result of the file to be audited according to the monomer elements and the fine-grained elements.
10. An electronic device, characterized in that: include: A memory for storing program instructions; A processor is used to call the program instructions stored in the memory and execute the letter of credit examination method according to any one of claims 1 to 8 according to the obtained program instructions.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a computer, the computer executes the letter of credit examination method as described in any one of claims 1 to 8.
12. A computer program product, characterized in that The computer program product comprises: a computer program code, which, when executed on a computer, enables the computer to execute the letter of credit examination method as claimed in any one of claims 1 to 8.