Method for parsing letter of credit text, method and device for reviewing letter of credit object

Abstract

The present invention discloses a method for parsing a letter of credit text, a method and device for reviewing letter of credit objects, and relates to the field of knowledge graph technology. A specific implementation of the method includes: obtaining a target letter of credit text, the target letter of credit text includes at least one target element; obtaining a target semantic template corresponding to the target letter of credit text, the target semantic template includes: a statement template and an upper shared hierarchy of the statement template, the statement template includes at least one target slot, and the target slot corresponds to the target element; according to the statement template and the upper shared hierarchy, the full-link slot of the target letter of credit text is obtained; the target semantic template is matched with the target letter of credit text to obtain the target slot value corresponding to the full-link slot. This implementation can automatically obtain the slots and slot values ​​corresponding to the elements in the letter of credit, and parse out the relevant content in the letter of credit text.

Description

Technical Field

[0001] The present invention relates to the field of knowledge graph technology, and in particular to a method for parsing letter of credit text, and a method and device for reviewing letter of credit objects. Background Art

[0002] A letter of credit is a written document issued by a bank to an exporter at the request of the importer, guaranteeing payment for goods. During the letter of credit review process, the bank compares the letter of credit requirements with the elements presented in each document to ensure consistency, as well as the consistency between each document. This process results in a "compliance" or "non-compliance" verdict. Traditional letter of credit review requires manual interpretation of the relevant content within the letter of credit or the letter of credit documents, as well as the corresponding values ​​for the relevant elements. This process is labor-intensive and inefficient. Summary of the Invention

[0003] In view of this, an embodiment of the present invention provides a method for parsing a letter of credit text, a method and device for reviewing a letter of credit object. The system can automatically obtain the slot and slot value corresponding to the target element in the letter of credit, thereby identifying the relevant content in the letter of credit text, reducing the manpower cost required in the letter of credit review process and increasing efficiency.

[0004] In a first aspect, an embodiment of the present invention provides a method for parsing a letter of credit text, comprising:

[0005] obtaining a target letter of credit text, wherein the target letter of credit text includes at least one target element;

[0006] Obtaining a target semantic template corresponding to the target letter of credit text, the target semantic template comprising: a sentence template and an upper shared layer of the sentence template, the sentence template comprising at least one target slot, the target slot corresponding to the target element;

[0007] Obtaining a full-link slot of the target letter of credit text according to the statement template and the upper shared level;

[0008] The target semantic template is matched with the target letter of credit text to obtain a target slot value corresponding to the full-link slot.

[0009] Optionally, the upper sharing level includes: a first operator;

[0010] The step of obtaining the full-link slot of the target letter of credit text according to the statement template and the upper shared level includes:

[0011] Determining, according to the first operator, a first combination of the upper sharing level and the target slot;

[0012] In the first combination method, the upper shared level and the target slot are combined to obtain the full-link slot of the target letter of credit text.

[0013] Optionally, the upper sharing level includes: a first sharing level, a second sharing level and a second operator;

[0014] Combining the upper shared level and the target slot in the first combination manner to obtain the full-link slot of the target letter of credit text includes:

[0015] Combining the first sharing level and the target slot in the first combination manner to obtain a first link slot;

[0016] Combining the second sharing level and the target slot in the first combination manner to obtain a second link slot;

[0017] determining, according to the second operator, a second combination of the first link slot and the second link slot;

[0018] In the second combination mode, the first link slot and the second link slot are combined to obtain the full link slot of the target letter of credit text.

[0019] Optionally, the second operator is located in the upper shared hierarchical part of the target semantic template, and the second operator includes at least one of the following: and, or, and output mode.

[0020] Optionally, the first operator includes at least one of the following: no intention, single intention, or intention and output mode.

[0021] Optionally, the statement template includes: a first target slot, a second target slot, and a third operator;

[0022] The step of obtaining the full-link slot of the target letter of credit text according to the statement template and the upper shared level includes:

[0023] combining the upper shared level and the first target slot to obtain a third link slot;

[0024] combining the upper shared layer and the second target slot to obtain a fourth link slot;

[0025] determining, according to the third operator, a third combination of the third link slot and the fourth link slot;

[0026] In the third combination mode, the third link slot and the fourth link slot are combined to obtain the full link slot of the target letter of credit text.

[0027] Optionally, the third operator is located in the sentence template part of the target semantic template, and the third operator includes at least one of the following: and, or, and replacement.

[0028] In a second aspect, an embodiment of the present invention provides a method for reviewing a letter of credit object, comprising:

[0029] Obtaining a first letter of credit text of a first letter of credit object, and using the above-mentioned letter of credit text parsing method to obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot;

[0030] Obtaining a second letter of credit text of the second letter of credit object, and using the above-mentioned letter of credit text parsing method to obtain a second text slot of the second letter of credit text and a slot value corresponding to the second text slot;

[0031] Determine whether the first letter of credit object is consistent with the second letter of credit object according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

[0032] Optionally, the first letter of credit text corresponds to at least one first text slot;

[0033] The determining whether the first letter of credit object and the second letter of credit object are consistent with each other according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot includes:

[0034] For each of the first text slots: determining a current second text slot corresponding to the current first text slot; respectively determining a first slot value corresponding to the current first text slot and a second slot value corresponding to the current second text slot; determining whether the first slot value and the second slot value match; if they do not match, determining that the current first text slot is an error slot;

[0035] Determining whether the first text slot includes an error slot;

[0036] If included, determining that the first letter of credit object is inconsistent with the second letter of credit object;

[0037] If not included, it is determined that the first letter of credit object is consistent with the second letter of credit object.

[0038] In a third aspect, an embodiment of the present invention provides a device for parsing a letter of credit text, comprising:

[0039] a text acquisition module, configured to acquire a target letter of credit text, wherein the target letter of credit text includes at least one target element;

[0040] a template acquisition module, configured to acquire a target semantic template corresponding to the target letter of credit text, wherein the target semantic template includes: a sentence template and an upper shared layer of the sentence template, wherein the sentence template includes at least one target slot, and the target slot corresponds to the target element;

[0041] A slot determination module, configured to obtain a full-link slot of the target letter of credit text according to the statement template and the upper shared hierarchy;

[0042] A value determination module is used to match the target semantic template with the target letter of credit text to obtain the target slot value corresponding to the full-link slot.

[0043] Optionally, the upper sharing level includes: a first operator;

[0044] The slot determination module is specifically configured to:

[0045] Determining, according to the first operator, a first combination of the upper sharing level and the target slot;

[0046] In the first combination method, the upper shared level and the target slot are combined to obtain the full-link slot of the target letter of credit text.

[0047] In a fourth aspect, an embodiment of the present invention provides a device for reviewing a letter of credit object, comprising:

[0048] a first acquisition module, configured to acquire a first letter of credit text of a first letter of credit object, and obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot using the above-mentioned letter of credit text parsing method;

[0049] a second acquisition module, configured to acquire a second letter of credit text of the second letter of credit object, and obtain a second text slot of the second letter of credit text and a slot value corresponding to the second text slot using the above-mentioned letter of credit text parsing method;

[0050] The review module is used to determine whether the first letter of credit object is consistent with the second letter of credit object according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

[0051] In a fifth aspect, an embodiment of the present invention provides an electronic device, including:

[0052] one or more processors;

[0053] a storage device for storing one or more programs,

[0054] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of the above embodiments.

[0055] In a sixth aspect, an embodiment of the present invention provides a computer-readable medium having a computer program stored thereon, which implements the method described in any of the above embodiments when the program is executed by a processor.

[0056] In a seventh aspect, an embodiment of the present invention provides a computer program product, including a computer program, which implements the method described in any of the above embodiments when executed by a processor.

[0057] One embodiment of the above invention has the following advantages or beneficial effects: according to the sentence template and the upper shared level in the target semantic template corresponding to the target letter of credit text, the full-link slot of the target letter of credit text is obtained. The full-link slot is the slot that needs to be extracted from the target letter of credit text. Then the target semantic template is matched with the target letter of credit text to obtain the target slot value corresponding to the full-link slot. Therefore, the relevant content in the letter of credit text can be automatically identified, the labor cost required in the letter of credit review process is reduced, and the efficiency is high.

[0058] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings are provided for a better understanding of the present invention and are not intended to limit the present invention.

[0060] Figure 1 1 is a flow chart of a method for parsing a letter of credit provided by the first embodiment of the present invention;

[0061] Figure 2 1 is a flow chart of a method for parsing a letter of credit text provided by the second embodiment of the present invention;

[0062] Figure 3 1 is a flow chart of a method for parsing a letter of credit provided by a third embodiment of the present invention;

[0063] Figure 4 1 is a flow chart of a method for reviewing a letter of credit object provided by a fourth embodiment of the present invention;

[0064] Figure 5 This is a schematic structural diagram of a device for parsing a letter of credit provided by an embodiment of the present invention;

[0065] Figure 6 This is a schematic structural diagram of a device for reviewing a letter of credit object provided by an embodiment of the present invention;

[0066] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0068] The acquisition, storage, use, and processing of data in this application's technical solution comply with relevant national laws and regulations.

[0069] To facilitate understanding of the solutions of the embodiments of the present invention, the following explains the terms involved in the embodiments of the present invention:

[0070] Knowledge representation: A subcategory of knowledge graph that describes the organizational form of knowledge and is the cornerstone of knowledge graph application.

[0071] Intent: The core content of the sentence, such as "Play me Jay Chou's Blue and White Porcelain" has the intention of "listening to music".

[0072] Slot: Entity category, defined as the category of related content items, such as the slots "singer" and "song title".

[0073] Slot value: that is, entity, which is the specific value of the slot in the current statement, such as the slot value "Jay Chou (singer)" and "Blue and White Porcelain (song name)".

[0074] Corresponding to the application of natural language understanding, understanding the terms of a letter of credit can be expressed at the following four levels:

[0075] (1) Identify the intent of the clause, such as "prohibit certain conduct" or "must display elements";

[0076] (2) Identify the slot under this intention in the clause, which is generally the name of the element, such as "Letter of Credit Number", "Signatory", etc.;

[0077] (3) Identify the specific slot value under the joint requirement of the intent and slot in the clause, which is generally a string or a numeric value;

[0078] (4) Assemble the above intentions, slots, and slot values ​​according to the knowledge representation structure defined by the knowledge graph to express the complete content of the clause requirements.

[0079] The embodiment of the present invention represents the semantics of clauses based on the knowledge representation structure defined by the knowledge graph, which is similar to the general knowledge representation definition and has the characteristics of multi-level inheritance relationship and sharing of basic element structure.

[0080] Elements correspond to slots and are the basic units for semantic recognition or parsing. Elements can be the name of the shipper, the name of the shipper's country, the name of the beneficiary, the beneficiary's phone number, etc.<subject,property,object> The following are examples of the basic elements:

[0081] <Party,hasProperty,name> Elements used to represent the shipper's name;

[0082] <Party,hasProperty,address> Elements used to represent the shipper's address;

[0083] <Party,hasProperty,country> An element used to represent the name of the country stated by the shipper.

[0084] Use "." to connect the upper and lower levels of the elements, and "(*)" to describe the content of the layer. The complete element (slot) is as follows:

[0085] Table 1 Examples of some data attribute elements of letters of credit

[0086] Chinese instructions Element (slot) Slot Type Requires shipper's name on bill of lading billLadingRequiredClause.shipper.name DP Requires the shipper's country name on the bill of lading billLadingRequiredClause.shipper.country.name DP Request shipper's phone number on bill of lading billLadingRequiredClause.shipper.contactInfo.phoneNumber DP … Require beneficiary name on invoice invoiceRequiredClause.beneficiary.name DP Require the name of the beneficiary's country on the invoice invoiceRequiredClause.beneficiary.country.name DP Request beneficiary's phone number on invoice invoiceRequiredClause.beneficiary.contactInfo.phoneNumber DP

[0087] In Table 1, DP indicates that the slot is a data property. Furthermore, if the terms of a letter of credit only require the shipper or beneficiary to be listed on the document without specifying their specific name, the corresponding element (slot) is OP (object property).

[0088] Table 2 Examples of some object attributes of letters of credit

[0089] Chinese instructions Element (slot) Slot Type Require shipper on bill of lading billLadingRequiredClause.shiper OP Requires the shipper's country on the bill of lading billLadingRequiredClause.shipper.country OP Require beneficiary name on invoice invoiceRequiredClause.beneficiary OP Request beneficiary information on invoices invoiceRequiredClause.beneficiary.contactInfo OP

[0090] To facilitate the identification of letter of credit text, long clauses within the text need to be broken down into their individual elements. In the closed language context of letters of credit, business elements are densely expressed, and most element units are independent. Therefore, embodiments of the present invention propose breaking long clauses down into their individual elements, parsing them separately according to their semantics, and then merging them into a complete semantic structure based on combination rules. Long clauses can be broken down into their own independent element units.

[0091] Example 1.SIGNED COMMERCIAL INVOICE IN THREE COPIES INDICATEMANUFACTURER'S / EXPORTOR'S NAME AND ADDRESS,LC NO.ABC123456 AND DT 2020 MAY20.

[0092] Example 1 can be broken down into the following independent elements, each separated by a semicolon: SIGNED; COMMERCIAL INVOICE; IN THREE COPIES; INDICATE MANUFACTURER'S / EXPORTOR'S NAME AND ADDRESS; LC NO.ABC123456 AND DT 2020 MAY 20.

[0093] Example 2.SIGNED COMMERCIAL INVOICE OR PACKING LIST SHOW LC NUMBER ANDDATE,GOODS AMOUNT,PRICE TERM.

[0094] Example 2 can be broken down into the following independent elements, each separated by a semicolon: SIGNED; COMMERCIAL INVOICE OR PACKING LIST SHOW; LC NUMBER AND DATE; GOODS AMOUNT; PRICE TERM.

[0095] Example 3.IN CASE OF BY COURIER SHIPMENT,BENE'S CERT NOT REQUIRED.

[0096] Sentence 3 can be broken down into the following independent elements, each separated by a semicolon: IN CASE OF BYCOURIER SHIPMENT; BENE'S CERT NOT REQUIRED.

[0097] As the above example demonstrates, different element units fully express the required semantic information and are independent of each other. Elements may contain logical relationships such as AND, OR, and IN CASE OF, and their effects vary depending on the element value. (If the document name contains logical relationships, the logical relationship applies to the entire sentence; if other elements contain logical relationships, the logic is limited to that element.)

[0098] Based on the above analysis, an embodiment of the present invention provides a method for parsing a letter of credit text. Figure 1: is a flowchart of a method for parsing a letter of credit text provided by the first embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0099] Step 101: Obtain a target letter of credit text, wherein the target letter of credit text includes at least one target element.

[0100] The target letter of credit text is the text related to the letter of credit. The target letter of credit text may be a text extracted from the letter of credit or the letter of credit document.

[0101] Elements correspond to slots and are the basic units for semantic recognition or parsing. Elements can include shipper name, shipper country name, beneficiary name, beneficiary phone number, etc.

[0102] The target letter of credit text can be a short text. The target letter of credit text involved in the embodiments of the present invention is a long text that can be split into multiple independent short texts for parsing. A short text includes one element or multiple related elements, and the short text can be parsed using a single semantic template, thereby achieving the purpose of parsing the entire long text.

[0103] Step 102: Acquire a target semantic template corresponding to the target letter of credit text, wherein the target semantic template includes a sentence template and an upper shared layer of the sentence template.

[0104] The sentence template includes at least one target slot, which corresponds to the target element. By matching the sentence template with the target letter of credit text, the slot value corresponding to the slot can be determined. A specific example of a semantic template is given below:

[0105] {address: <address>consignee

[0106] In the above formula, the left side of ">>" is the statement template part, and the right side is the upper shared level part of the slot. The semantic template part may include one or more target slots. The upper shared level part may include one or more upper shared levels. The upper shared level is generally a None or OP attribute. The upper shared level can be simplified by "intention", and the description of the embodiment of the present invention will also replace the upper shared level with intention.

[0107] Step 103: Obtain the full-link slot of the target letter of credit text according to the statement template and the upper sharing level.

[0108] A full-link slot is a slot in the target letter of credit text that needs to be parsed for a slot value. A full-link slot corresponds to the target element in the target letter of credit text. Semantic operators can be included in semantic templates. These operators determine how the statement template is combined with the upper-level shared hierarchy to generate the full-link slot for the target letter of credit text.

[0109] Step 104: Match the target semantic template with the target letter of credit text to obtain the target slot value corresponding to the full-link slot.

[0110] In an embodiment of the present invention, the full-link slot of the target letter of credit text is obtained based on the sentence template in the target semantic template corresponding to the target letter of credit text and the upper shared hierarchy. The full-link slot is the slot that needs to be extracted from the target letter of credit text. The target semantic template is then matched with the target letter of credit text to obtain the target slot value corresponding to the full-link slot. Therefore, the relevant content in the letter of credit text can be automatically identified, reducing the labor cost required in the letter of credit review process and being more efficient.

[0111] Figure 2 This is a flow chart of a method for parsing a letter of credit text provided by the second embodiment of the present invention. Figure 2 As shown, the method includes:

[0112] Step 201: Obtain a target letter of credit text, wherein the target letter of credit text includes at least one target element.

[0113] Step 202: Acquire a target semantic template corresponding to the target letter of credit text, wherein the target semantic template includes: a statement template and an upper shared layer of the statement template, wherein the upper shared layer includes: a first operator.

[0114] The first operator is used to determine a first combination of an upper shared level and a target slot. The first operator includes at least one of the following: no intent, a single intent, or an intent and output mode.

[0115] Step 203: Determine a first combination of the upper sharing level and the target slot according to the first operator.

[0116] The first combination mode includes at least one of the following: a non-intentional mode, a single intentional mode, or an intentional mode and an output mode.

[0117] Step 204: Combine the upper shared level and the target slot in the first combination manner to obtain the full-link slot of the target letter of credit text.

[0118] Step 205: Match the target semantic template with the target letter of credit text to obtain the target slot value corresponding to the full-link slot.

[0119] In an embodiment of the present invention, according to the first operator in the upper sharing layer, the first combination method of the upper sharing layer and the target slot is determined, and then according to the first combination method, the upper sharing layer and the target slot are combined, so that the full-link slot of the target letter of credit text can be obtained flexibly and conveniently.

[0120] In one embodiment of the present invention, the upper sharing layer includes: a first sharing layer, a second sharing layer and a second operator; combining the upper sharing layer and the target slot in the first combination method to obtain the full link slot of the target letter of credit text includes: combining the first sharing layer and the target slot in the first combination method to obtain the first link slot; combining the second sharing layer and the target slot in the first combination method to obtain the second link slot; determining the second combination method of the first link slot and the second link slot according to the second operator; combining the first link slot and the second link slot in the second combination method to obtain the full link slot of the target letter of credit text.

[0121] The second operator is located in the upper shared layer of the target semantic template and can be used to connect multiple upper shared layers. The second operator includes at least one of the following output methods: AND, OR, and. The second combination method includes at least one of the following output methods: AND, OR, and output. The first and second operators can be used to flexibly and conveniently obtain the full-link slot of the target letter of credit document.

[0122] In one embodiment of the present invention, the statement template includes: a first target slot, a second target slot and a third operator; obtaining the full link slot of the target letter of credit text based on the statement template and the upper shared layer includes: combining the upper shared layer and the first target slot to obtain a third link slot; combining the upper shared layer and the second target slot to obtain a fourth link slot; determining a third combination method of the third link slot and the fourth link slot according to the third operator; and combining the third link slot and the fourth link slot in the third combination method to obtain the full link slot of the target letter of credit text.

[0123] The third operator is located in the statement template portion of the target semantic template. The third operator can be used to connect multiple target slots. The third operator includes at least one of the following: AND, OR, and REPLACE. The third combination includes at least one of the following: AND, OR, and REPLACE. Using the third operator, the full link slot of the target letter of credit text can be flexibly and conveniently obtained.

[0124] To facilitate understanding of the embodiments of the present invention, the following explains concepts such as semantic templates and semantic operators with specific examples. Among them, the semantic operators include: a first operator, a second operator, and a third operator. Based on the inheritance and sharing characteristics of the knowledge representation structure, and the semantic independence of the element units in long sentence clauses, the embodiment of the present invention proposes an element parsing rule based on semantic templates, and defines a set of semantic operators and their representation requirements.

[0125] Semantic template 1: IN? {totalCopy: <num>}[COPIES|COPY]>>None

[0126] Semantic template 2: [CONSIGNEED|CONSIGNEE]AND[NOTIFIED|NOTIFY]{name: <company>}[,|#]{address: <address>AND_consignee_notifyParty

[0127] In the above semantic templates 1 and 2, the left side of "">>" is the statement template, and the right side is the upper shared level of the slot, which is simplified with "intention" and generally has the None or OP attribute. "*?" indicates that the word can be present or absent; "{}" defines the finest-grained slot. The slot name is before ":" and the specific slot value is after it. "<*>" is a wildcard for a type of slot value, generally used for preprocessing results or scenarios where the dictionary can be replaced, such as <company>This corresponds to a company name library. "[*|*]" describes synonyms or other similar interchangeable expressions without affecting semantics, and is separated by vertical bars. "#" indicates that the word at that position can be empty, similar to the function of "?". Therefore, for the following long sentence to be parsed, its complete semantic meaning can be expressed using five semantic templates.

[0128] Long sentence to be parsed: SIGNED COMMERCIAL INVOICE IN THREE COPIES INDICATEMANUFACTURER'S / EXPORTOR'S NAME AND ADDRESS,LC NO.ABC123456 AND DT 2020 MAY20.

[0129] Semantic template 1: ALL? DULY? {show_signer:SIGNED} >> None

[0130] Semantic template 2: {docTitle: <document>}>>None

[0131] Semantic template 3: IN? {totalCopy: <num>}[COPIES|COPY]>>None

[0132] Semantic Template 4: ACTUAL?[MANUFACTURER'S|EXPORTER'S][OR| / ][MANUFACTURER'S|EXPORTER'S]{show_name:Name}[AND|,]{show_address:ADDRESS}>>OR_exporter_manufacturer

[0133] Semantic Template 5: [DOCUMENTARY CREDIT|LETTER OF CREDIT|LC|DC][NUMBER|NO.|NO]{creditNumber: <code>}[AND|,][DATED|DT|DD]{lcIssueDate: <date>}>>None

[0134] Assuming the above sentence belongs to the category invoiceRequiredClause, the semantics of the entire sentence are obtained by concatenating the "intent" on the right and the slots on the left with ".". The left side of the colon is the slot, and the right side of the colon is the slot value obtained by matching the sentence to be parsed using the semantic template:

[0135] invoiceRequiredClause.signer:”

[0136] invoiceRequiredClause.docTitle: COMMERCIAL INVOICE

[0137] invoiceRequiredClause.totalCopy: 3

[0138] (invoiceRequiredClause.exporter.name: "&invoiceRequiredClause.exporter.address: ") | (invoiceRequiredClause.manufacturer.name: "&invoiceRequiredClause.manufacturer.address: ")

[0139] invoiceRequiredClause.creditNumber: 123456

[0140] invoiceRequiredClause.lcIssueDate: 2020 MAY 20

[0141] Based on the elements and their relationships required by the letter of credit language, the following semantic operators (first operator, second operator, and third operator) are defined:

[0142] AND / OR comes first in intention or intent;

[0143] ">>OR" expresses the OR relationship of the left slot;

[0144] ">>AND / OR_A" expresses the intention that A and the left slot form an AND / OR relationship;

[0145] ">>AND / OR_A_B" expresses the intention that A and B are in an AND / OR relationship, and then spliced ​​with the left slot;

[0146] The or slot is placed first, indicating that multiple slots defined under or have an or relationship, such as "ISSUED BY{or_name: <company>}OR{or_name: <company>}>>issuer";

[0147] None intention, the clause content is directly expressed by the left slot;

[0148] show in the first place of intent / slot, converting the original non-boolean attribute to boolean intent

[0149] ">>show_A” indicates the document display slot A, and outputs {A:”};

[0150] ">>show_A_B" indicates that the values ​​of multiple slots A and B are Boolean and in an AND relationship;

[0151] ">>showOR_A_B" indicates that the values ​​of multiple slots A and B are Boolean and in an OR relationship;

[0152] replace_a@b indicates that the slot value refers to a scenario and needs to be replaced by the specific value of "b" after "@". For example, the output of {replace_issuer.name@seller.name} is {issuer.name:ABC LTD}, where "ABC LTD" is seller.name.

[0153] a_b indicates that multiple slots share one slot value. For example, the output of the slot {issuer.name_seller.name:ABC LTD} is {issuer.name:ABC LTD}, {seller.name:ABC LTD}.

[0154] The embodiment of the present invention proposes a complex clause parsing and structure restoration solution based on template combination. Figure 3 : is a flowchart of a method for parsing a letter of credit text provided by the third embodiment of the present invention, such as Figure 3 As shown, the target semantic template is input, and the target semantic template includes: a statement template and an upper shared layer of the statement template, and the statement template includes at least one target slot, and the upper shared layer can be used to represent the intent. Determine the subsequent splicing operation according to the form of the intent and the calculation contained in the semantic template. Traverse the slots, and splice with the intent according to the slot form and operator to obtain the full-link slot. According to the position of the OR operator, determine whether to extend the OR intent to the right clause. Use the target semantic template to match the target letter of credit text, and output the semantic information of the target letter of credit text. According to the semantic template and semantic operator defined above, the business meaning of the letter of credit terms can be restored according to the intent, slot, and slot value, and directly mapped to the knowledge representation of the knowledge graph. In the above different scenarios, the combination of intent and slots is different.

[0155] Table 3 Structure restoration logic based on intent slots and semantic operators

[0156]

[0157] Table 3 illustrates the structure reduction logic based on intent slots and semantic operators. Using the logical relationships shown in Table 3, semantic information can be parsed from the letter of credit text. A or B can be used to represent intent or slot. The intent in Table 3 can also be understood as a higher-level shared hierarchy.

[0158] In Table 3, the None intent corresponds to the no intent of the first operator. The single intent corresponds to the single intent of the first operator. The "OR" in the OR_intent corresponds to the OR mode of the third operator. The "OR" in OR_A corresponds to the OR intent of the first operator. The "OR" in OR_A_B corresponds to the OR mode of the second operator. The "AND" in AND_A_B corresponds to the AND mode of the second operator. The "show" in Show_A corresponds to the output mode of the first operator. The "show" in Show_A can also correspond to the output mode of the third operator. For showOR_A_B, if A and B are intents, then showOR corresponds to the output mode or intent of the first operator; if A and B are slots, then showOR corresponds to the output mode or OR mode of the third operator. The "replace" in Replace_A@B corresponds to the replacement mode of the third operator.

[0159] For ease of understanding, taking the Invoice Required Clause as an example, the semantic restoration results of each scenario are given (you can also refer to the semantic analysis and structural restoration results of the complete long sentence above):

[0160] Text A: SIGNED OR STAMPED BY BENEFICIARY

[0161] Template A: SIGNED OR STAMPED BY{partyIndicator:BENEFICIARY}>>OR_signer_stamper

[0162] The "OR" in "">>OR_signer_stamper" in template A corresponds to the second operator, OR. Using template A, we can extract the full link slots and slot values ​​from text A:

[0163] InvoiceRequiredClause.signer.partyIndicator: BENEFICIARY|InvoiceRequiredClause.stamper.partyIndicator: BENEFICIARY

[0164] Text B: SIGNED BY BENEFICIARY OR ABC LTD

[0165] Template B: ISSUED BY THE?{partyIndicator: BENEFICIARY}[OR| / ]{name: <company>}>>OR_issuer

[0166] The [OR| / ] in template B corresponds to the OR method of the third operator. The full link slot and slot value in text B are extracted through template B:

[0167] InvoiceRequiredClause.issuer.partyIndicator:BENEFICIARY|InvoiceRequiredClause.issuer.name:ABC LTD

[0168] Text C: LC Number and Date

[0169] Template C: [DOCUMENTARY CREDIT|LETTER OF CREDIT|LC|DC][NUMBER|NO.|NO][AND|,][ISSUING|ISSUED|#][DATE|DT|DD]>>show_creditNumber_lcIssueDate

[0170] The "show" in "show_creditNumber_lcIssueDate" in template C corresponds to the output mode of the second operator. The full link slot and slot value in text C are extracted using template C:

[0171] creditNumber:”&lcIssueDate:”

[0172] Text D: ISSUED BY THE SELLER

[0173] Template D:

[0174] ISSUED BY THE? {replace_name@seller.name:SELLER}>>issuer

[0175] The "issuer" in template D corresponds to the single intent of the first operator. The full link slots and slot values ​​in text D are extracted using template D:

[0176] issuer.name:ABC LTD, where "ABC LTD" is seller.name

[0177] Text E: INVOICE OR PACKING LIST MUST INDICATE GOODS ORIGIN.

[0178] Template E1: {docTitle: <document>}OR{originCountry: <origincountry>}>>OR

[0179] Template E2: [GOODS ORIGIN|ORIGIN COUNTRY|ORIGIN OF GOODS|COUNTRY OFORIGIN]>>show_originCountry

[0180] The first "OR" in template E1 corresponds to the OR method of the third operator. The "show" in template E2 corresponds to the output method of the first operator. Using templates E1 and E2, we can extract the full link slots and slot values ​​from text E:

[0181] (InvoiceRequiredClause.docTitle:INVOICE&InvoiceRequiredClause.originCountry:”)|(PackinglistRequiredClause.docTitle:PACKING LIST&PackinglistRequiredClause.originCountry:”)

[0182] Among them, texts A to D are short texts that can be parsed using a single semantic template. Text E is a long text that needs to be split into multiple independent short texts for parsing.

[0183] The language of letters of credit has the problems of high element density, many logical scenarios, and large differences in slot types, making it difficult to use methods based on deep learning models (the sample demand is extremely large, and the slot definition is difficult). Therefore, an embodiment of the present invention proposes a parsing method for letter of credit text based on a combination of semantic templates, in which the relationships between multiple elements and logical sentences are all reflected in short templates, which greatly reduces the complexity of parsing. The embodiment of the present invention also proposes a set of semantic algorithms suitable for letter of credit language, which can cover scenarios such as various attributes, attribute booleanization, and attribute logical combination, further avoiding the situation where attribute combination scenarios cannot be handled in traditional slot definitions. The combination of semantic operators and slots cleverly designed in the solution of the embodiment of the present invention can cover richer semantic representations, avoiding the problem that slots in traditional parsing can only be redundantly defined by scenario.

[0184] In addition, the embodiment of the present invention proposes a "short sentence meaning template parsing" method to split the difficult problem of parsing long sentences and complex letter of credit requirements into sub-problems of parsing each short sentence, thereby reducing the difficulty of parsing. In addition, through a cleverly designed set of semantic operators, it is possible to cover complex attribute value recognition, Booleanization of non-Boolean attributes ("show_A" form or "show_A_B" form, etc.), shared attribute hypernyms, recognition of complex attributes with logical AND / OR, etc., avoiding the redundancy and complexity of defining slots for each type of attribute in traditional slot identification. Finally, through the operation process corresponding to the semantic operator, the short sentence parsing results are combined to obtain the semantic representation of the complete long sentence and complex terms. Therefore, the solution of the embodiment of the present invention has the following technical effects:

[0185] (1) The long sentence parsing is decoupled into short sentence parsing according to its independent semantic meaning, and a parsing expression is designed that can cover synonym replacement, optional words, slot extraction, and intention (slot sharing hypernym).

[0186] (2) A set of semantic operators has been designed to cover slot identification problems in special scenarios such as various attributes, attribute booleanization, and attribute logical combination. In addition, the semantic operators are bound to their operation processes and can be combined according to the combination logic to restore the complete semantics of the long sentence.

[0187] (3) By combining slots and operators, the automatic recognition of various complex semantics of the letter of credit text can be achieved.

[0188] Figure 4 FIG. 4 is a flow chart of a method for reviewing a letter of credit object provided by the fourth embodiment of the present invention. Figure 4 As shown, the method includes:

[0189] Step 401: Obtain a first letter of credit text of a first letter of credit object, and use the above-mentioned letter of credit text parsing method to obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot.

[0190] The first letter of credit object is an object related to the letter of credit. The first letter of credit object can be a letter of credit, a letter of credit document, a letter of credit clause, etc. The semantics of the first letter of credit text can be parsed using the first text slot and the slot value corresponding to the first text slot.

[0191] Step 402: Obtain the second letter of credit text of the second letter of credit object, and use the above-mentioned letter of credit text parsing method to obtain the second text slot of the second letter of credit text and the slot value corresponding to the second text slot.

[0192] The second letter of credit object is an object related to the letter of credit. The second letter of credit object can be a letter of credit, a letter of credit document, a letter of credit clause, etc. The semantics of the second letter of credit text can be parsed using the second text slot and the slot value corresponding to the second text slot.

[0193] Step 403: Determine whether the first letter of credit object is consistent with the second letter of credit object based on the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

[0194] Specifically, whether the first letter of credit object is consistent with the second letter of credit object can be determined in the following manner: for each of the first text slots: determine the current second text slot corresponding to the current first text slot; determine the first slot value corresponding to the current first text slot and the second slot value corresponding to the current second text slot respectively; determine whether the first slot value matches the second slot value; if not, determine that the current first text slot is an error slot; determine whether the first text slot contains an error slot; if so, determine that the first letter of credit object is inconsistent with the second letter of credit object; if not, determine that the first letter of credit object is consistent with the second letter of credit object.

[0195] By using the above method, the first slot value corresponding to each first slot in the first LC object is compared with the second slot value corresponding to the second slot. If any first slot value does not match the second slot value, it indicates that the first LC object and the second LC object are inconsistent. If all first slot values ​​have corresponding second slot values, it indicates that the first LC object and the second LC object are consistent.

[0196] The method for parsing the letter of credit text provided in the embodiment of the present invention is applied to the letter of credit review. The system can automatically obtain the slots and slot values ​​corresponding to the elements in the letter of credit object, thereby identifying the semantic content contained in the letter of credit object, reducing the manpower cost required in the letter of credit review process and improving efficiency.

[0197] Figure 5 This is a schematic diagram of the structure of a parsing device for a letter of credit provided by an embodiment of the present invention. Figure 5 As shown, the device includes:

[0198] A text acquisition module 501 is used to acquire a target letter of credit text, wherein the target letter of credit text includes at least one target element;

[0199] A template acquisition module 502 is configured to acquire a target semantic template corresponding to the target letter of credit text, wherein the target semantic template includes a sentence template and an upper shared layer of the sentence template, wherein the sentence template includes at least one target slot corresponding to the target element;

[0200] A slot determination module 503 is configured to obtain a full-link slot of the target letter of credit text according to the statement template and the upper shared hierarchy;

[0201] The value determination module 504 is used to match the target semantic template with the target letter of credit text to obtain the target slot value corresponding to the full-link slot.

[0202] Optionally, the upper sharing level includes: a first operator;

[0203] The slot determination module 503 is specifically configured to:

[0204] Determining, according to the first operator, a first combination of the upper sharing level and the target slot;

[0205] In the first combination method, the upper shared level and the target slot are combined to obtain the full-link slot of the target letter of credit text.

[0206] Optionally, the second operator includes at least one of the following: a sum, an or an output mode.

[0207] Optionally, the first operator includes at least one of the following: no intention, single intention, or intention and output mode.

[0208] Optionally, the statement template includes: a first target slot, a second target slot, and a third operator;

[0209] The slot determination module 503 is specifically configured to:

[0210] combining the upper shared level and the first target slot to obtain a third link slot;

[0211] combining the upper shared layer and the second target slot to obtain a fourth link slot;

[0212] determining, according to the third operator, a third combination of the third link slot and the fourth link slot;

[0213] In the third combination mode, the third link slot and the fourth link slot are combined to obtain the full link slot of the target letter of credit text.

[0214] Optionally, the third operator includes at least one of the following: and, or, and replacement.

[0215] Figure 6 This is a schematic diagram of the structure of a device for reviewing a letter of credit object provided by an embodiment of the present invention. Figure 6 As shown, the device includes:

[0216] A first acquisition module 601 is configured to acquire a first letter of credit text of a first letter of credit object, and obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot using the above-mentioned letter of credit text parsing method;

[0217] A second acquisition module 602 is configured to acquire a second letter of credit text of a second letter of credit object, and obtain a second text slot of the second letter of credit text and a slot value corresponding to the second text slot using the above-mentioned letter of credit text parsing method;

[0218] The review module 603 is configured to determine whether the first letter of credit object is consistent with the second letter of credit object based on the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

[0219] Optionally, the first letter of credit text corresponds to at least one first text slot;

[0220] The audit module 603 is specifically used for:

[0221] For each of the first text slots: determining a current second text slot corresponding to the current first text slot; respectively determining a first slot value corresponding to the current first text slot and a second slot value corresponding to the current second text slot; determining whether the first slot value and the second slot value match; if they do not match, determining that the current first text slot is an error slot;

[0222] Determining whether the first text slot includes an error slot;

[0223] If included, determining that the first letter of credit object is inconsistent with the second letter of credit object;

[0224] If not included, it is determined that the first letter of credit object is consistent with the second letter of credit object.

[0225] An embodiment of the present invention provides an electronic device, including:

[0226] one or more processors;

[0227] a storage device for storing one or more programs,

[0228] When one or more programs are executed by one or more processors, the one or more processors implement the method of any of the above embodiments.

[0229] An embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the method for parsing a letter of credit text or the method for reviewing a letter of credit object in the embodiment of the present invention.

[0230] Reference below Figure 7 , which shows a schematic structural diagram of a computer system 700 of a terminal device suitable for implementing an embodiment of the present invention. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0231] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage unit 708 into a random access memory (RAM) 703. Various programs and data required for the operation of the system 700 are also stored in the RAM 703. The CPU 701, ROM 702, and RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0232] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, and the like; an output section 707 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 708 including a hard disk; and a communication section 709 including a network interface card such as a LAN card or a modem. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 710 as needed, so that computer programs read therefrom can be installed into the storage section 708 as needed.

[0233] In particular, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 709, and / or installed from a removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0234] It should be noted that the computer-readable medium described in the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media can include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber cable, RF, or any suitable combination thereof.

[0235] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0236] The modules described in the embodiments of the present invention may be implemented via software or hardware. These modules may also be located within a processor. For example, they may be described as a text acquisition module, a template acquisition module, a slot determination module, and a value determination module. In some cases, the names of these modules do not limit the modules themselves. For example, the text acquisition module may be described as "a module for acquiring a target letter of credit text, wherein the target letter of credit text includes at least one target element."

[0237] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiments, or may exist independently without being incorporated into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the device, the device includes:

[0238] obtaining a target letter of credit text, wherein the target letter of credit text includes at least one target element;

[0239] Obtaining a target semantic template corresponding to the target letter of credit text, the target semantic template comprising: a sentence template and an upper shared layer of the sentence template, the sentence template comprising at least one target slot, the target slot corresponding to the target element;

[0240] Obtaining a full-link slot of the target letter of credit text according to the statement template and the upper shared level;

[0241] The target semantic template is matched with the target letter of credit text to obtain a target slot value corresponding to the full-link slot.

[0242] According to the technical solution of an embodiment of the present invention, the full-link slot of the target letter of credit text is obtained based on the sentence template and the upper shared hierarchy in the target semantic template corresponding to the target letter of credit text. The full-link slot is the slot that needs to be extracted from the target letter of credit text. The target semantic template is then matched with the target letter of credit text to obtain the target slot value corresponding to the full-link slot. Therefore, the relevant content in the letter of credit text can be automatically identified, reducing the labor cost required in the letter of credit review process and being more efficient.

[0243] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.< / origincountry> < / document> < / company> < / company> < / company> < / date> < / code> < / num> < / document> < / company> < / address> < / company> < / num> < / address>

Claims

1. A method for parsing a letter of credit, characterized in that: include: obtaining a target letter of credit text, wherein the target letter of credit text includes at least one target element; Obtaining a target semantic template corresponding to the target letter of credit text, the target semantic template comprising: a statement template and an upper shared layer of the statement template, the statement template comprising at least one target slot, the target slot corresponding to the target element; wherein the upper shared layer comprises: a first operator; Obtaining a full-link slot of the target letter of credit text according to the statement template and the upper-level shared hierarchy, comprising: determining a first combination method of the upper-level shared hierarchy and the target slot according to the first operator; and combining the upper-level shared hierarchy and the target slot in the first combination method to obtain the full-link slot of the target letter of credit text; The target semantic template is matched with the target letter of credit text to obtain a target slot value corresponding to the full-link slot.

2. The method according to claim 1, characterized in that The upper sharing layer includes: a first sharing layer, a second sharing layer and a second operator; Combining the upper shared level and the target slot in the first combination manner to obtain the full-link slot of the target letter of credit text includes: Combining the first sharing level and the target slot in the first combination manner to obtain a first link slot; Combining the second sharing level and the target slot in the first combination manner to obtain a second link slot; determining, according to the second operator, a second combination of the first link slot and the second link slot; In the second combination mode, the first link slot and the second link slot are combined to obtain the full link slot of the target letter of credit text.

3. The method according to claim 2, characterized in that The second operator is located in the upper shared layer part of the target semantic template, and the second operator includes at least one of the following: an AND operator, an OR operator, and an output mode.

4. The method according to claim 1, wherein The first operator includes at least one of the following: no intention, single intention, or intention and output mode.

5. The method according to claim 1, wherein The statement template includes: a first target slot, a second target slot and a third operator; The step of obtaining the full-link slot of the target letter of credit text according to the statement template and the upper shared level includes: combining the upper shared level and the first target slot to obtain a third link slot; combining the upper shared level and the second target slot to obtain a fourth link slot; determining, according to the third operator, a third combination of the third link slot and the fourth link slot; In the third combination mode, the third link slot and the fourth link slot are combined to obtain the full link slot of the target letter of credit text.

6. The method according to claim 5, characterized in that The third operator is located in the sentence template part of the target semantic template, and the third operator includes at least one of the following: and, or, and replacement.

7. A method for reviewing a letter of credit object, characterized in that: include: Obtaining a first letter of credit text of a first letter of credit object, using the method according to any one of claims 1 to 6, to obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot; Obtaining a second letter of credit text of the second letter of credit object, using the method according to any one of claims 1 to 6, to obtain a second text slot of the second letter of credit text and a slot value corresponding to the second text slot; Determine whether the first letter of credit object is consistent with the second letter of credit object according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

8. The method according to claim 7, characterized in that The first letter of credit text corresponds to at least one first text slot; The determining whether the first letter of credit object and the second letter of credit object are consistent with each other according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot includes: For each of the first text slots: determining a current second text slot corresponding to the current first text slot; respectively determining a first slot value corresponding to the current first text slot and a second slot value corresponding to the current second text slot; determining whether the first slot value and the second slot value match; if they do not match, determining that the current first text slot is an error slot; Determining whether the first text slot includes an error slot; If included, determining that the first letter of credit object is inconsistent with the second letter of credit object; If not included, it is determined that the first letter of credit object is consistent with the second letter of credit object.

9. A device for parsing a letter of credit, characterized in that: include: a text acquisition module, configured to acquire a target letter of credit text, wherein the target letter of credit text includes at least one target element; A template acquisition module is configured to acquire a target semantic template corresponding to the target letter of credit text, wherein the target semantic template includes: a statement template and an upper shared layer of the statement template; the statement template includes at least one target slot, and the target slot corresponds to the target element; wherein the upper shared layer includes: a first operator; A slot determination module is configured to obtain a full-link slot of the target letter of credit text based on the statement template and the upper-level shared hierarchy, comprising: determining a first combination method of the upper-level shared hierarchy and the target slot based on the first operator; and combining the upper-level shared hierarchy and the target slot using the first combination method to obtain the full-link slot of the target letter of credit text; A value determination module is used to match the target semantic template with the target letter of credit text to obtain the target slot value corresponding to the full-link slot.

10. A credit card object review device, characterized in that: include: a first acquisition module, configured to acquire a first letter of credit text of a first letter of credit object, using the method according to any one of claims 1 to 6 to obtain a first text slot of the first letter of credit text and a slot value corresponding to the first text slot; a second obtaining module, configured to obtain a second letter of credit text of a second letter of credit object, using the method according to any one of claims 1 to 6 to obtain a second text slot of the second letter of credit text and a slot value corresponding to the second text slot; The review module is used to determine whether the first letter of credit object is consistent with the second letter of credit object according to the slot value corresponding to the first text slot and the slot value corresponding to the second text slot.

11. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

12. A computer readable medium having a computer program stored thereon, characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

13. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images