CN117635045A - Intelligent receipt contract management method, device and system - Google Patents

Intelligent receipt contract management method, device and system Download PDF

Info

Publication number
CN117635045A
CN117635045A CN202311616536.1A CN202311616536A CN117635045A CN 117635045 A CN117635045 A CN 117635045A CN 202311616536 A CN202311616536 A CN 202311616536A CN 117635045 A CN117635045 A CN 117635045A
Authority
CN
China
Prior art keywords
instruction
matching degree
user input
contract management
instruction set
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202311616536.1A
Other languages
Chinese (zh)
Inventor
李兴华
简志枰
李红霞
陈铭洙
吴兴威
伍慧彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Construction Bank Corp
CCB Finetech Co Ltd
Original Assignee
China Construction Bank Corp
CCB Finetech Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Construction Bank Corp, CCB Finetech Co Ltd filed Critical China Construction Bank Corp
Priority to CN202311616536.1A priority Critical patent/CN117635045A/en
Publication of CN117635045A publication Critical patent/CN117635045A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/103—Workflow collaboration or project management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/38—Payment protocols; Details thereof
    • G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
    • G06Q20/405—Establishing or using transaction specific rules

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Finance (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Machine Translation (AREA)

Abstract

The embodiment of the application provides an intelligent receipt contract management method, device and system, and relates to the technical field of intelligent receipt. The intelligent receipt contract management method comprises the following steps: constructing an instruction operation conversion component based on the AIGC model, wherein the instruction operation conversion component is used for converting instructions in a first instruction set into operation steps in a second instruction set through the trained AIGC model; acquiring a user input instruction; respectively calculating a first matching degree of a user input instruction and a first instruction set and a second matching degree of the user input instruction and a second instruction set; selecting different conversion modes according to the first matching degree and the second matching degree to convert the user input instruction into an operation step; the obtained operation steps are output to the order receiving contract management system, and the return result of the order receiving contract management system is used as a response to the input instruction of the user. The embodiment provided by the application improves the processing efficiency of the receipt contract and also improves the compatibility of the input instruction.

Description

Intelligent receipt contract management method, device and system
Technical Field
The present application relates to the technical field of intelligent order receiving, and in particular, to an intelligent order receiving contract management method, an intelligent order receiving contract management device, an intelligent order receiving contract management system, a computer readable storage medium, and a computer program product.
Background
The order receiving service refers to the action that an order receiving mechanism signs a bank card acceptance protocol with an special-about merchant, and after the special-about merchant accepts the bank card according to the specification and deals with a cardholder, the special-about merchant is provided with a transaction fund settlement service. Along with the great increase of special merchants, massive acquiring contract information is generated, and the acquiring mechanism needs to maintain corresponding resources input by the acquiring mechanism, so that higher labor cost and system cost are caused.
At present, contract management personnel of a receiving mechanism can carry out corresponding operation in a receiving contract management system only by inputting corresponding maintenance instructions, the operation skill requirements of the management personnel are high, and when a large number of operations are carried out, operation errors are easy to generate, so that the efficiency is low. Moreover, the current order-receiving contract management system has low intelligent degree and cannot adapt to the modern management requirements of an order-receiving mechanism.
AIGC (Artificial Intelligence Generated Content): generating artificial intelligence.
Disclosure of Invention
The purpose of the embodiment of the application is to provide an intelligent order-receiving contract management method, device and system, which utilize AIGC technology to help an order-receiving row to manage huge order-receiving contract information by using relatively low labor cost, so that contract management staff of an order-receiving mechanism can operate in the intelligent order-receiving contract management system through simple voice or text description instructions, and at least part of technical problems in the background technology are solved.
To achieve the above object, a first aspect of the present application provides an intelligent order contract management method, including: constructing an instruction operation conversion component based on the AIGC model, wherein the instruction operation conversion component is used for converting instructions in a first instruction set into operation steps in a second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system; acquiring a user input instruction; respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set; when the second matching degree is larger than a preset second matching degree threshold value, mapping the user input instruction into a second instruction set; when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component; and outputting the obtained operation steps to the order receiving contract management system, and taking a return result of the order receiving contract management system as a response to the user input instruction.
Preferably, the trained AIGC model is obtained by: acquiring a historical operation data set; the data in the historical operation data set comprises a mapping relation between the first instruction set and the second instruction set; performing data cleaning and data calibration on the historical operation data set to obtain processed data; training the AIGC model by adopting the processed data as a training sample; and when the preset training ending condition is reached, obtaining the trained AIGC model.
Preferably, according to the relation between the first matching degree and a preset first matching degree threshold, selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set, including: when the first matching degree is larger than a preset first matching degree threshold value, normalizing the user input instruction based on the grammar structure of the first instruction set to obtain an instruction in the first instruction set; and when the first matching degree is not greater than a preset first matching degree threshold value, identifying semantic elements in the user input instruction, and reconstructing the user input instruction into an instruction in a first instruction set based on the semantic elements.
Preferably, identifying a semantic element in the user input instruction, reconstructing the user input instruction into an instruction in a first instruction set based on the semantic element, includes: acquiring a text corresponding to the user input instruction; if the input instruction is in a non-text mode, converting the input instruction into a corresponding text; performing lexical analysis and/or grammar analysis on the text to obtain predicates and objects in the text; if the modifier and/or the qualifier also exist, determining the position of the modifier and/or the qualifier according to the modifier and/or the qualifier; adjusting the predicate, object, modifier, qualifier based on the vocabulary of the first instruction set; reconstructing the adjusted predicates, objects, modifiers, qualifiers as instructions in the first instruction set based on the syntax structure of the first instruction set.
Preferably, the method further comprises: setting a third matching degree threshold value for matching the first matching degree, wherein the third matching degree threshold value is lower than the first matching degree threshold value; and outputting a prompt which cannot be identified by the instruction as a response to the user input instruction when the second matching degree is not greater than a preset second matching degree threshold and the first matching degree is lower than the third matching degree threshold.
Preferably, the first instruction set is constructed based on the steps of: determining a plurality of term sets based on operation elements in an order contract management system, and generating a term table based on the plurality of term sets; performing word segmentation recognition and screening on the collected related documents of the acquiring contract management system according to the glossary to obtain a keyword set in the acquiring contract management system; and combining the keyword sets through a natural language-like grammar structure to obtain the first instruction set.
Preferably, determining a plurality of term sets based on operation elements in the order contract management system, and generating a term table based on the plurality of term sets, includes: generating a predicate set based on an operation instruction in the order receiving contract management system; generating an object set based on the operation object in the order contract management system; generating a modifier set and/or a qualifier set based on the switch parameter of the operation instruction and the attribute parameter of the operation object; the glossary is generated based on the set of predicate, the set of object, the set of modifier, and the set of qualifiers.
Preferably, the term list is used for identifying and screening the collected related documents of the acquiring contract management system in terms to obtain a keyword set in the acquiring contract management system, and the keyword set comprises the following components: performing word segmentation recognition on the collected related documents of the acquiring contract management system according to the glossary to obtain a word segmentation recognition result; the word frequency of the word segmentation recognition result is subjected to inverted indexing, and then a high-frequency term is obtained through a word cloud picture; and filtering redundant data in the high-frequency term to obtain a keyword set.
Preferably, the method further comprises updating the first instruction set by: acquiring updated related documents of the acquiring contract management system; performing word segmentation recognition on the updated related document to obtain a word segmentation recognition result; processing word frequency of the segmentation word recognition result by using the following punishment function to obtain punished word frequency: w=c×1/(V), where W is the word frequency after penalty, C is the word frequency before penalty, and V is the average word frequency of the word in multiple historical iterations; selecting new keywords not in the keyword set based on the punished word frequency; and updating the first instruction set according to the new key word.
Preferably, when the second matching degree is greater than a preset second matching degree threshold, the operation step of mapping the user input instruction into a second instruction set includes: after judging that the second matching degree is larger than a preset second matching degree threshold value, checking whether the current user has model modification authority; and when the current user has the model modification right, mapping the user input instruction into a second instruction set.
In a second aspect of the present application, there is also provided an intelligent order contract management apparatus, the apparatus including: the conversion component construction module is used for constructing an instruction operation conversion component based on the AIGC model, and the instruction operation conversion component is used for converting the instructions in the first instruction set into operation steps in the second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system; the input acquisition module is used for acquiring a user input instruction; the matching calculation module is used for calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set respectively; the first mapping module is used for mapping the user input instruction into an operation step in a second instruction set when the second matching degree is larger than a preset second matching degree threshold value; the second mapping module is used for selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set according to the relation between the first matching degree and a preset first matching degree threshold when the second matching degree is not greater than the preset second matching degree threshold, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component; and the execution response module is used for outputting the obtained operation steps to the order receiving contract management system and taking a return result of the order receiving contract management system as a response to the user input instruction.
In a third aspect of the present application, there is also provided an intelligent order contract management system, including: the system comprises a user input component, an instruction operation conversion component, a model training component and a receipt contract management component; the instruction operation conversion component is used for converting the instructions in the first instruction set into operation steps in the second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system; the user input component is used for: acquiring an input instruction of a user; respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set; when the second matching degree is larger than a preset second matching degree threshold value, mapping the user input instruction into a second instruction set; when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component; the order contract management component is used for executing the operation steps; the model training component is configured to train an AIGC model in the instruction manipulation transformation component.
In a fourth aspect of the present application, there is also provided an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the foregoing intelligent acquirement contract management method when executing the computer program.
In a fifth aspect of the present application, there is also provided a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the steps of the aforementioned intelligent acquirement contract management method.
In a sixth aspect the present application provides a computer program product comprising a computer program which when executed by a processor implements the aforementioned intelligent acquirer contract management method.
The technical scheme has at least the following beneficial effects:
(1) The artificial intelligence technology is introduced, so that the processing efficiency of the contract for acquiring the list is obviously improved, and the labor cost of contract management is reduced;
(2) The system processes the input instruction in various modes, and improves the intellectualization of the receipt contract processing.
(3) And a plurality of parameter modification modes are supported, so that the flexibility of the order-receiving contract processing system is improved.
Additional features and advantages of embodiments of the present application will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the present application and are incorporated in and constitute a part of this specification, illustrate embodiments of the present application and together with the description serve to explain, without limitation, the embodiments of the present application. In the drawings:
FIG. 1 schematically illustrates a step schematic diagram of an intelligent acquirer contract management method according to an embodiment of the present application;
FIG. 2 schematically illustrates a process diagram from natural language to an operation set in accordance with an embodiment of the present application;
FIG. 3 schematically illustrates an overall process flow diagram of a system according to an embodiment of the present application;
FIG. 4 schematically illustrates a step schematic of an AIGC model training process according to an embodiment of the application;
FIG. 5 schematically illustrates a structural diagram of an intelligent acquirer contract management apparatus according to an embodiment of the present application;
FIG. 6 schematically illustrates a structural diagram of an intelligent acquirer contract management system according to an embodiment of the present application;
fig. 7 schematically shows an internal structural view of a computer device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the specific embodiments described herein are only for illustrating and explaining the embodiments of the present application, and are not intended to limit the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
In the present embodiment, if directional indications (such as up, down, left, right, front, and rear … …) are included, the directional indications are merely used to explain the relative positional relationship, movement, and the like between the components in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indications are correspondingly changed.
In addition, if there is a description of "first", "second", etc. in the embodiments of the present application, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the protection scope of the present application.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations.
Fig. 1 schematically illustrates a step diagram of an intelligent acquirer contract management method according to an embodiment of the present application. As shown in fig. 1, an intelligent receipt contract management method includes:
s01, constructing an instruction operation conversion component based on an AIGC model, wherein the instruction operation conversion component is used for converting instructions in a first instruction set into operation steps in a second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system;
s02, acquiring a user input instruction;
s03, respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set;
s04, mapping the user input instruction into a second instruction set when the second matching degree is larger than a preset second matching degree threshold value;
s04', when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component;
S05, outputting the obtained operation steps to the order receiving contract management system, and taking a return result of the order receiving contract management system as a response to the user input instruction.
Through the above embodiment, besides introducing AIGC technology into order-receiving contract management to improve management efficiency and reduce labor cost, the input of the AIGC model is preprocessed by setting the first instruction set, so that the form of model input instructions is expanded. Meanwhile, the method can convert the user input instruction into the operation steps of the order receiving contract management system by adopting different methods based on the matching degree, so that the limitation of the input format is avoided, the use scene of the system is obviously expanded, and the user experience is improved. In the embodiment, the operation steps of direct input are optimized, so that AIGC model conversion is skipped, and processing time delay is reduced.
In some embodiments of the present application, it should be noted that step S01 is constructed in advance, and the intelligent receipt contract management method is executed from step S02 when it is in the actual application scenario. The aforementioned step numbers merely illustrate the interrelationship between the steps, and step S01 need not be performed each time before step S02.
The names of the first instruction set and the second instruction set in the embodiments in this application are only used to distinguish the two, and the names are not limited thereto. For example, hereinafter, based on the functions of the first instruction set, which is also referred to as an order contract management instruction set, and the second instruction set, which is also referred to as an order contract management operation set. In addition, how to calculate the matching degree in the foregoing embodiment may refer to a matching degree calculation manner in the prior art, which is not described herein.
The step S04 is mainly applied to the following scenarios: after the instruction conversion model is generated, both the first instruction set and the second instruction set are changed due to the development of the receiving management service. When the model training assembly is changed, the change can be realized after multiple times of training. Thus, at the time of change, the "model training component" supports changes with weights, i.e., the changes can be immediately reflected into the mapping model. For example, the default deduction rate of merchants on the policy-based line needs to be increased, and then the default templates used in the operation step need to be immediately modified and validated. To this end, the user input component of the present embodiment also supports direct adaptation of the model itself, i.e. direct conversion to an operational step. For example, by speaking "modify the default deduction rate of online merchants", the user input component first checks whether the user has the authority to modify the model, and if so, generates instructions to modify the online merchant template. The instructions may be embodied as a set of operations that modify the modified click through values, ultimately acting on the model.
In some embodiments of the present application, according to a relationship between a first matching degree and a preset first matching degree threshold, selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set includes: when the first matching degree is larger than a preset first matching degree threshold value, normalizing the user input instruction based on the grammar structure of the first instruction set to obtain an instruction in the first instruction set; and when the first matching degree is not greater than a preset first matching degree threshold value, identifying semantic elements in the user input instruction, and reconstructing the user input instruction into an instruction in a first instruction set based on the semantic elements. After the foregoing relative relationship between the second matching degree and the preset second matching degree threshold is determined, it can be determined that the user input instruction does not belong to the operation step in the second instruction set, and at this time, it is required to further determine whether the user input instruction belongs to or is similar to the instruction in the first instruction set. When the first matching degree is greater than a preset first matching degree threshold value, the user input instruction may be determined to be an instruction in the first instruction set or similar to an instruction in the first instruction set. Only a grammatical legitimacy check of the user input instruction is required. And when the grammar legitimacy is not available, the grammar structure is adjusted, so that the user input instruction belongs to the instruction in the first instruction set. In another case, when the first matching degree is not greater than the preset first matching degree threshold, the user input instruction may be a natural language instruction, and larger conversion is needed, at this time, the lexical analyzer and the grammar analyzer are adopted to analyze, extract semantic elements therein, normalize the semantic elements, and then recombine the normalized semantic elements into an instruction in the first instruction set.
In an alternative embodiment of the present application, a semantic element based reconstruction step is provided, specifically including: acquiring a text corresponding to the user input instruction; if the input instruction is in a non-text mode, converting the input instruction into a corresponding text; for example, when the user inputs an instruction as speech, it is necessary to convert the instruction into text by a speech recognition technique. Performing lexical analysis and/or grammar analysis on the text to obtain predicates and objects in the text; if the modifier and/or the qualifier also exist, determining the position of the modifier and/or the qualifier according to the modifier and/or the qualifier; adjusting the predicate, object, modifier, qualifier based on the vocabulary of the first instruction set; the purpose of the adjustment is to normalize the term. Reconstructing the adjusted predicates, objects, modifiers, qualifiers as instructions in the first instruction set based on the syntax structure of the first instruction set. If the grammar structure of the first instruction set is a natural language-like grammar structure, the adjusted predicates, objects, modifiers and qualifiers should be recombined by adopting the natural language-like grammar structure, so that the obtained combination meets the specification requirement of the first instruction set. When the words and grammar are both constructed according to the first instruction set, the obtained instructions naturally also belong to the instructions in the first instruction set.
In some embodiments of the present application, the method further comprises: setting a third matching degree threshold value for matching the first matching degree, wherein the third matching degree threshold value is lower than the first matching degree threshold value; and outputting a prompt which cannot be identified by the instruction as a response to the user input instruction when the second matching degree is not greater than a preset second matching degree threshold and the first matching degree is lower than the third matching degree threshold. In this embodiment, an abnormal situation is considered, that is, when the matching degree between the user input instruction and the second instruction set is insufficient and the matching degree between the user input instruction and the first instruction set is too low, the user input instruction is a failed instruction, and the system does not process the failed instruction and returns an abnormality, so that the safety of the system is improved, and malfunction of the system is prevented from being notified by input of an irrelevant instruction.
In some embodiments of the present application, there is further provided a step of constructing a first instruction set, including: determining a plurality of term sets based on operation elements in order contract management, and generating a term table based on the plurality of term sets; performing word segmentation recognition and screening on the collected related documents of the acquiring contract management system according to the glossary to obtain a keyword set in acquiring contract management; and combining the keyword sets through a natural language-like grammar structure to obtain the order-receiving contract management instruction set.
The first instruction set constructed by the above embodiment has mainly the following advantages: the order receiving contract management instruction set is a preset instruction set, is similar to natural language, follows a preset grammar structure, and is mainly used for carrying out certain standardization on the input of an instruction conversion model based on an artificial intelligent model so as to improve the recognition probability and the conversion accuracy. The advantage of this setting still lies in: when the AIGC technique is used to generate the pre-training model in the subsequent step, the pre-training model can be directly mixed with natural language, and the input set is not required to be a strict grammar instruction set.
In an alternative embodiment of the present application, determining a plurality of term sets based on an operation element in order contract management, and generating a term table based on the plurality of term sets includes: generating a predicate set based on the operation instruction in the order-receiving contract management; generating an object set based on the operation object in the order-receiving contract management; generating a modifier set and/or a qualifier set based on the switch parameter of the operation instruction and the attribute parameter of the operation object; the glossary is generated based on the set of predicate, the set of object, the set of modifier, and the set of qualifiers. Illustratively, the operation instructions in order taking contract management include a new instruction, a modification instruction, a deletion instruction, etc., and thus the corresponding predicate set includes a new addition, a modification, and a deletion. The object set is generated primarily from the operational objects, which include, but are not limited to, merchants, terminals, stores, deductions, jurisdictions, accounting institutions, and the like. In some cases, the operation instructions are provided with on-off parameters to achieve more personalized functions, or the operation objects are provided with attribute parameters, for example, the attribute parameters of the merchants can comprise on-line or off-line, so that the merchants can be divided into on-line merchants and off-line merchants based on the attribute parameters to perform operations on required parts of the merchants. There is therefore a need to add a set of modifiers and/or a set of qualifiers to achieve the customization and personalization of instructions.
In an optional embodiment of the present application, the word segmentation recognition and screening are performed on the collected related documents of the acquiring contract management system according to the glossary, so as to obtain a keyword set in acquiring contract management, including: performing word segmentation recognition on the collected related documents of the acquiring contract management system according to the glossary to obtain a word segmentation recognition result; the word frequency of the word segmentation recognition result is subjected to inverted indexing, and then a high-frequency term is obtained through a word cloud picture; and filtering redundant data in the high-frequency term to obtain a keyword set. Specifically, the instruction set for acquiring contract management in this embodiment is a set of special instruction sets including acquiring contract management technical term sets and operation keywords and natural language-like grammar structures. The generation process of the instruction set comprises the following steps: in order to improve the accuracy of word segmentation, before word segmentation, a glossary managed by a receipt contract is manually manufactured, the glossary is imported into a custom dictionary of a word segmentation plug-in unit, and the word segmentation plug-in unit can perform word segmentation recognition on the terms. Collecting all required documents, detailed designs, database designs and other design documents, operation and maintenance documents, operation manuals, problem sheets and other materials of the system, and carrying out word frequency analysis of natural language on the materials. The word frequency analysis process is to use Chinese word segmentation tool to segment the above materials, and count word frequency by reverse index after word segmentation. There are many tools currently used to accomplish the word segmentation and word frequency statistics, including elastiscearch (IK plugin for chinese word segmentation), reidsearch, etc. After word segmentation, inverted indexes are needed to be carried out on all word segmentation results of materials, and after the inverted indexes are created, related high-frequency terms can be simply checked by establishing a word cloud picture. The inverted results may contain redundant data, such as common words, which need to be manually removed to obtain a cleaner set of keywords.
An order contract management instruction set is created based on the set of keywords. The created order-receiving contract management instruction set is a set of instruction set special for the system. The instruction set is similar to natural language, and has the advantages that when the AIGC technology is used for generating a pre-training model, the instruction set can be directly mixed with the natural language, and an input set is not required to be a strict grammar instruction set. The method is characterized by comprising the following steps: 1. the lexical structure follows predicate and object structures. Such as a [ modify merchant XXX ] instruction in which "modify" is predicate and "merchant XXX" is object. 2. Predicates include new additions, modifications, deletions. 3. Objects include, but are not limited to: merchants, terminals, store, deduction rates, jurisdictions, accounting institutions, and the like. 4. The object may be preceded by modifier and qualifier, such as adding online merchants, modifying the deduction rate of XXX merchants.
In an embodiment of the present application, the operation step of converting the input instruction of the user into the order contract management system using the order contract management instruction set as an intermediary includes: normalizing an input instruction of a user into a text instruction, and converting the text instruction into a corresponding instruction in the order receiving contract management instruction set based on a text analysis model; and inputting the corresponding instruction in the order receiving contract management instruction set into an instruction conversion model based on the artificial intelligence model, and converting the instruction into operation steps in the order receiving contract management operation set, wherein the operation steps can be identified and executed by the order receiving contract management system. The present embodiment provides a method for generating operation steps of two conversions mediated mainly by an order-receiving contract management instruction set. The input instruction of the user can be voice, text or other various forms.
Fig. 2 schematically illustrates a process diagram from natural language to an operation set according to an embodiment of the present application. As shown in fig. 2, taking an example of a natural language instruction input by a user, a special instruction set, that is, an order-receiving contract management instruction set is obtained through analysis by a special lexical analyzer and analysis by a special grammar analyzer. And finally, converting the special instruction set into an order-receiving contract management operation set. Wherein the user's input instructions may take a variety of forms, such as: user voice input, text input, and specific order contract management instruction set input. When the input is voice, the user input component converts the voice into text input through a voice recognition technology; when the input is text input, the component can identify keywords in the order contract management instruction set; when the input is a receipt management instruction set, the component can check the validity of the input grammar.
In an optional embodiment of the present application, normalizing an input instruction of a user into a text instruction, converting the text instruction into a corresponding instruction in the order-receiving contract management instruction set based on a text analysis model, including: acquiring a text corresponding to the user input instruction; if the input instruction is in a non-text mode, converting the input instruction into a corresponding text; performing lexical analysis and/or grammar analysis on the text to obtain predicates and objects in the text; if the modifier and/or the qualifier also exist, determining the position of the modifier and/or the qualifier according to the modifier and/or the qualifier; and matching the predicates, the objects, the modifiers and the qualifiers in the order receiving contract management instruction set, and determining corresponding instructions from the order receiving contract management instruction set according to the similarity. The adopted models or methods of word segmentation, text acquisition and matching in the step can adopt the models or methods in the prior art, or can adopt a specially developed special lexical analyzer to perform lexical analysis, or a specially developed special grammar analyzer to perform grammar analysis, so as to translate into an operation set which can be understood by an order-receiving contract management system. For example, the user input information is "newly added online merchant", and after analysis and interpretation by the lexical analyzer and the grammar analyzer, the result is: the "new" is predicate "on-line merchant" is object, where "on-line" is a qualifier for the merchant. This "online" qualifier would correspond to "order contract management system operation" creating a merchant using an online merchant's template.
Fig. 3 schematically shows an overall process flow diagram of a system according to an embodiment of the present application. As shown in fig. 3, the overall system flow is as follows: the user inputs the demand description into the user input component, the user input component converts the demand description into an instruction set to be input into the instruction operation conversion component, the instruction operation conversion component consults the conversion result from the model training component instruction set to the executable operation set, the instruction operation conversion component converts the instruction set into the executable operation set to be input into the order collection contract management component, and the order collection contract management component executes the operation set to change contract information.
In an alternative embodiment of the present application, the order contract management instruction set is updated by: acquiring updated related documents of the acquiring contract management system; performing word segmentation recognition on the updated related document to obtain a word segmentation recognition result; processing word frequency of the segmentation word recognition result by using the following punishment function to obtain punished word frequency: w=c×1/(V), where W is the word frequency after penalty, C is the word frequency before penalty, and V is the average word frequency of the word in multiple historical iterations; selecting new keywords not in the keyword set based on the punished word frequency; and updating the order receiving contract management instruction set according to the new keyword. Specifically, as time goes on, new functions are introduced into the order taking contract management system, and the order taking contract management instruction set also needs to be updated correspondingly. The updating mode can adopt a construction mode of an order receiving contract management instruction set to carry out loop iteration so as to keep the latest state and the optimal performance. During loop iteration, some vocabularies with strong specialization such as "merchant", "contract", "store", "deduction rate" and the like and high universality occupy the front of high-frequency vocabularies for a long time, so that some new vocabularies cannot be revealed. These long-term front-holding keywords need to be penalized at this time by penalty functions. The penalty function used in this embodiment is:
W＝C*1/(V)；
Wherein W represents punished word frequency, C represents word frequency of the current iteration, and V represents average word frequency of the word or key word in the previous iterations. After punishment of word frequency, word frequency of the technical terms is lowered, new keywords appearing in new demands are found, the keywords generate order-receiving contract management instructions according to the method, and the order-receiving contract management instructions are added into an original order-receiving contract management instruction set.
In an alternative embodiment of the present application, the instruction transformation model is a generative artificial intelligence model, AIGC. Because the instruction set is a specialized condensed representation of the user's needs, when the user's needs are implemented in the receiving management system, the instruction set needs to be implemented as a series of operation steps (i.e., an operation set) for the receiving management system to recognize and execute. In the prior art, the mapping process from an instruction set to an operation set is realized by a fixed rule mapping or a manual intervention mapping mode, and the mapping process has the defects of low flexibility or high labor cost. The present embodiment employs an AIGC technique, so that the mapping process from the instruction set to the operation set can be automatically completed. The automatic mapping generation process is completed by the instruction operation conversion component by utilizing the mapping rules provided by the model training component. According to the embodiment, the revolutionary AIGC technology is adopted, and the AIGC technology is applied to merchant contract management, so that huge order-receiving contract information is managed by using relatively low labor cost, and contract management personnel of an order-receiving mechanism can input a simple voice or text description instruction into an intelligent order-receiving contract management system, so that the order-receiving contract is managed.
The training process of the generated artificial intelligence model is as follows: acquiring a historical operating dataset of the instruction operation conversion component; the historical operation data set comprises a mapping relation between the order receiving contract management instruction set and the order receiving contract management operation set; performing data cleaning and data calibration on the historical operation data set to obtain processed data; and the model training component adopts the processed data as a training sample and trains the instruction conversion model. In this embodiment, the model training component trains out an instruction conversion model that can be automatically mapped from instruction set to operation set using AIGC technique.
FIG. 4 schematically illustrates a step diagram of an AIGC model training process according to an embodiment of the application, as illustrated in FIG. 4, the training process of the instruction conversion model is: the aforementioned instruction operation conversion component records a historical operation data set, i.e. the mapping relationship from instruction set to operation set. Because the duration of the data acquisition is long, for example, it may last years. Over a longer period of time, both the instruction set and the operation set of the merchant may change, for example, certain fields are added or subtracted from the acquirer contract information, and default values in the acquirer contract information template have changed. These varying data can have a large impact on training results. To mask these effects, data cleaning and data calibration are required for the acquired data. The data cleaning function is to clean out data that is not actually available. The data calibration serves to update data that can be corrected by partial modification. The data set after data cleaning and data calibration can be used as input data of a training model. The training process is performed with reference to the prior art and will not be described in detail here.
Through the steps in the above embodiments, complex instruction conversion is set inside the model, and implemented by using AIGC. The professional requirement on the input instruction is reduced, and the processing efficiency is improved.
Based on the same inventive concept, the application also provides an intelligent receipt contract management device. Fig. 5 schematically illustrates a structural diagram of an intelligent receipt contract management apparatus according to an embodiment of the present application. As shown in fig. 5, the conversion component construction module is configured to construct an instruction operation conversion component based on the AIGC model, where the instruction operation conversion component is configured to convert an instruction in a first instruction set into an operation step in a second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system; the input acquisition module is used for acquiring a user input instruction; the matching calculation module is used for calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set respectively; the first mapping module is used for mapping the user input instruction into an operation step in a second instruction set when the second matching degree is larger than a preset second matching degree threshold value; the second mapping module is used for selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set according to the relation between the first matching degree and a preset first matching degree threshold when the second matching degree is not greater than the preset second matching degree threshold, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component; and the execution response module is used for outputting the obtained operation steps to the order receiving contract management system and taking a return result of the order receiving contract management system as a response to the user input instruction.
In some alternative embodiments, the trained AIGC model is obtained by: acquiring a historical operation data set; the data in the historical operation data set comprises a mapping relation between the first instruction set and the second instruction set; performing data cleaning and data calibration on the historical operation data set to obtain processed data; training the AIGC model by adopting the processed data as a training sample; and when the preset training ending condition is reached, obtaining the trained AIGC model.
In some optional embodiments, according to a relationship between the first matching degree and a preset first matching degree threshold, selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set includes: when the first matching degree is larger than a preset first matching degree threshold value, normalizing the user input instruction based on the grammar structure of the first instruction set to obtain an instruction in the first instruction set; and when the first matching degree is not greater than a preset first matching degree threshold value, identifying semantic elements in the user input instruction, and reconstructing the user input instruction into an instruction in a first instruction set based on the semantic elements.
In some alternative embodiments, identifying semantic elements in the user input instruction, reconstructing the user input instruction into instructions in a first instruction set based on the semantic elements, includes: acquiring a text corresponding to the user input instruction; if the input instruction is in a non-text mode, converting the input instruction into a corresponding text; performing lexical analysis and/or grammar analysis on the text to obtain predicates and objects in the text; if the modifier and/or the qualifier also exist, determining the position of the modifier and/or the qualifier according to the modifier and/or the qualifier; adjusting the predicate, object, modifier, qualifier based on the vocabulary of the first instruction set; reconstructing the adjusted predicates, objects, modifiers, qualifiers as instructions in the first instruction set based on the syntax structure of the first instruction set.
In some alternative embodiments, the apparatus further comprises: setting a third matching degree threshold value for matching the first matching degree, wherein the third matching degree threshold value is lower than the first matching degree threshold value; and outputting a prompt which cannot be identified by the instruction as a response to the user input instruction when the second matching degree is not greater than a preset second matching degree threshold and the first matching degree is lower than the third matching degree threshold.
In some alternative embodiments, the first instruction set is constructed based on the steps of: determining a plurality of term sets based on operation elements in an order contract management system, and generating a term table based on the plurality of term sets; performing word segmentation recognition and screening on the collected related documents of the acquiring contract management system according to the glossary to obtain a keyword set in the acquiring contract management system; and combining the keyword sets through a natural language-like grammar structure to obtain the first instruction set.
In some alternative embodiments, determining a plurality of term sets based on an operational element in an order contract management system, and generating a term table based on the plurality of term sets, includes: generating a predicate set based on an operation instruction in the order receiving contract management system; generating an object set based on the operation object in the order contract management system; generating a modifier set and/or a qualifier set based on the switch parameter of the operation instruction and the attribute parameter of the operation object; the glossary is generated based on the set of predicate, the set of object, the set of modifier, and the set of qualifiers.
In some optional embodiments, the term list is used for identifying and filtering the collected related documents of the acquiring contract management system in a word segmentation way, so as to obtain a keyword set in the acquiring contract management system, which comprises the following steps: performing word segmentation recognition on the collected related documents of the acquiring contract management system according to the glossary to obtain a word segmentation recognition result; the word frequency of the word segmentation recognition result is subjected to inverted indexing, and then a high-frequency term is obtained through a word cloud picture; and filtering redundant data in the high-frequency term to obtain a keyword set.
In some alternative embodiments, the apparatus further comprises updating the first instruction set by: acquiring updated related documents of the acquiring contract management system; performing word segmentation recognition on the updated related document to obtain a word segmentation recognition result; processing word frequency of the segmentation word recognition result by using the following punishment function to obtain punished word frequency: w=c×1/(V), where W is the word frequency after penalty, C is the word frequency before penalty, and V is the average word frequency of the word in multiple historical iterations; selecting new keywords not in the keyword set based on the punished word frequency; and updating the first instruction set according to the new key word.
In some optional embodiments, when the second matching degree is greater than a preset second matching degree threshold, the operation step of mapping the user input instruction into a second instruction set includes: after judging that the second matching degree is larger than a preset second matching degree threshold value, checking whether the current user has model modification authority; and when the current user has the model modification right, mapping the user input instruction into a second instruction set.
The specific definition of each functional module in the above-mentioned intelligent receipt contract management device may refer to the definition of each step in the intelligent receipt contract management method, which is not described herein. The various modules in the foregoing system may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules. The device acts on the field of order contract management, and has the advantage of improving the efficiency of order contract management.
The application also provides an intelligent receipt contract management system. Fig. 6 schematically illustrates a structural diagram of an intelligent order contract management system according to an embodiment of the present application. As shown in fig. 6, the system includes: the system comprises a user input component, an instruction operation conversion component, a model training component and a receipt contract management component; the instruction operation conversion component is used for converting the instructions in the first instruction set into operation steps in the second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system; the user input component is used for: acquiring an input instruction of a user; respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set; when the second matching degree is larger than a preset second matching degree threshold value, mapping the user input instruction into a second instruction set; when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component; the order contract management component is used for executing the operation steps; the model training component is configured to train an AIGC model in the instruction manipulation transformation component. The intelligent order-receiving contract management system focuses on the application side, and the first instruction set is integrated in the user input component.
The embodiment of the application provides a storage medium, and a program is stored on the storage medium, and when the program is executed by a processor, the intelligent receipt contract management method is realized.
In one embodiment, a computer device is provided, which may be a server, and the internal structure thereof may be as shown in fig. 7. Fig. 7 schematically shows an internal structural view of a computer device according to an embodiment of the present application. The computer device includes a processor a01, a network interface a02, a memory (not shown) and a database (not shown) connected by a system bus. Wherein the processor a01 of the computer device is adapted to provide computing and control capabilities. The memory of the computer device includes internal memory a03 and nonvolatile storage medium a04. The nonvolatile storage medium a04 stores an operating system B01, a computer program B02, and a database (not shown in the figure). The internal memory a03 provides an environment for the operation of the operating system B01 and the computer program B02 in the nonvolatile storage medium a04. The network interface a02 of the computer device is used for communication with an external terminal through a network connection. The computer program B02, when executed by the processor a01, implements an intelligent order contract management method.
It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, the intelligent acquirer contract management system provided herein may be implemented in the form of a computer program that is executable on a computer device as shown in FIG. 7. The memory of the computer device may store therein respective program modules constituting the intelligent order contract management system, and the computer program constituted by the respective program modules causes the processor to execute the steps in the intelligent order contract management method of the respective embodiments of the present application described in the present specification.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer-readable media include both permanent and non-permanent, removable and non-removable media, and information storage may be implemented by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
The foregoing is merely an embodiment of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims (14)

1. An intelligent order-receiving contract management method is characterized by comprising the following steps:
constructing an instruction operation conversion component based on the AIGC model, wherein the instruction operation conversion component is used for converting instructions in a first instruction set into operation steps in a second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system;
Acquiring a user input instruction;
respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set;
when the second matching degree is larger than a preset second matching degree threshold value, mapping the user input instruction into a second instruction set;
when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component;
and outputting the obtained operation steps to the order receiving contract management system, and taking a return result of the order receiving contract management system as a response to the user input instruction.
2. The method of claim 1, wherein the trained AIGC model is obtained by:
acquiring a historical operation data set; the data in the historical operation data set comprises a mapping relation between a first instruction set and a second instruction set;
Performing data cleaning and data calibration on the historical operation data set to obtain processed data;
training the AIGC model by adopting the processed data as a training sample;
and when the preset training ending condition is reached, obtaining the trained AIGC model.
3. The method of claim 1, wherein selecting the first conversion mode or the second conversion mode to convert the user input instruction into an instruction in the first instruction set according to a relationship between the first matching degree and a preset first matching degree threshold value comprises:
when the first matching degree is larger than a preset first matching degree threshold value, normalizing the user input instruction based on the grammar structure of the first instruction set to obtain an instruction in the first instruction set;
and when the first matching degree is not greater than a preset first matching degree threshold value, identifying semantic elements in the user input instruction, and reconstructing the user input instruction into an instruction in a first instruction set based on the semantic elements.
4. A method according to claim 3, wherein identifying semantic elements in the user input instruction, reconstructing the user input instruction based on the semantic elements as an instruction in a first instruction set, comprises:
Acquiring a text corresponding to the user input instruction; if the input instruction is in a non-text mode, converting the input instruction into a corresponding text;
performing lexical analysis and/or grammar analysis on the text to obtain predicates and objects in the text; if the modifier and/or the qualifier also exist, determining the position of the modifier and/or the qualifier according to the modifier and/or the qualifier;
adjusting the predicate, object, modifier, qualifier based on the vocabulary of the first instruction set;
reconstructing the adjusted predicates, objects, modifiers, qualifiers as instructions in the first instruction set based on the syntax structure of the first instruction set.
5. The method according to claim 1, wherein the method further comprises:
setting a third matching degree threshold value for matching the first matching degree, wherein the third matching degree threshold value is lower than the first matching degree threshold value;
and outputting a prompt which cannot be identified by the instruction as a response to the user input instruction when the second matching degree is not greater than a preset second matching degree threshold and the first matching degree is lower than the third matching degree threshold.
6. The method of claim 1, wherein the first instruction set is constructed based on the steps of:
Determining a plurality of term sets based on operation elements in an order contract management system, and generating a term table based on the plurality of term sets;
performing word segmentation recognition and screening on the collected related documents of the acquiring contract management system according to the glossary to obtain a keyword set in the acquiring contract management system;
and combining the keyword sets through a natural language-like grammar structure to obtain the first instruction set.
7. The method of claim 6, wherein determining a plurality of term sets based on the operational elements in the acquirer contract management system and generating a glossary based on the plurality of term sets comprises:
generating a predicate set based on an operation instruction in the order receiving contract management system;
generating an object set based on the operation object in the order contract management system;
generating a modifier set and/or a qualifier set based on the switch parameter of the operation instruction and the attribute parameter of the operation object;
the glossary is generated based on the set of predicate, the set of object, the set of modifier, and the set of qualifiers.
8. The method of claim 6, wherein the step of word segmentation recognition and screening of the collected related documents of the acquiring contract management system according to the glossary to obtain the keyword set in the acquiring contract management system comprises the steps of:
Performing word segmentation recognition on the collected related documents of the acquiring contract management system according to the glossary to obtain a word segmentation recognition result;
the word frequency of the word segmentation recognition result is subjected to inverted indexing, and then a high-frequency term is obtained through a word cloud picture;
and filtering redundant data in the high-frequency term to obtain a keyword set.
9. The method of claim 6, further comprising updating the first instruction set by:
acquiring updated related documents of the acquiring contract management system;
performing word segmentation recognition on the updated related document to obtain a word segmentation recognition result;
processing word frequency of the segmentation word recognition result by using the following punishment function to obtain punished word frequency:
w=c×1/(V), where W is the word frequency after penalty, C is the word frequency before penalty, and V is the average word frequency of the word in multiple historical iterations;
selecting new keywords not in the keyword set based on the punished word frequency;
and updating the first instruction set according to the new key word.
10. The method of claim 1, wherein the operation step of mapping the user input instructions into a second instruction set when the second degree of matching is greater than a preset second degree of matching threshold comprises:
After judging that the second matching degree is larger than a preset second matching degree threshold value, checking whether the current user has model modification authority;
and when the current user has the model modification right, mapping the user input instruction into a second instruction set.
11. An intelligent order contract management apparatus, the apparatus comprising:
the conversion component construction module is used for constructing an instruction operation conversion component based on the AIGC model, and the instruction operation conversion component is used for converting the instructions in the first instruction set into operation steps in the second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system;
the input acquisition module is used for acquiring a user input instruction;
the matching calculation module is used for calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set respectively;
the first mapping module is used for mapping the user input instruction into an operation step in a second instruction set when the second matching degree is larger than a preset second matching degree threshold value;
The second mapping module is used for selecting a first conversion mode or a second conversion mode to convert the user input instruction into an instruction in a first instruction set according to the relation between the first matching degree and a preset first matching degree threshold when the second matching degree is not greater than the preset second matching degree threshold, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component;
and the execution response module is used for outputting the obtained operation steps to the order receiving contract management system and taking a return result of the order receiving contract management system as a response to the user input instruction.
12. An intelligent order contract management system, comprising: the system comprises a user input component, an instruction operation conversion component, a model training component and a receipt contract management component;
the instruction operation conversion component is used for converting the instructions in the first instruction set into operation steps in the second instruction set through the trained AIGC model; the operation steps in the second instruction set are operation steps in the order-receiving contract management system;
the user input component is used for: acquiring an input instruction of a user; respectively calculating a first matching degree of the user input instruction and the first instruction set and a second matching degree of the user input instruction and the second instruction set; when the second matching degree is larger than a preset second matching degree threshold value, mapping the user input instruction into a second instruction set; when the second matching degree is not greater than a preset second matching degree threshold value, selecting a first conversion mode or a second conversion mode according to the relation between the first matching degree and the preset first matching degree threshold value to convert the user input instruction into an instruction in a first instruction set, and converting the obtained instruction into a corresponding operation step through the instruction operation conversion component;
The order contract management component is used for executing the operation steps;
the model training component is configured to train an AIGC model in the instruction manipulation transformation component.
13. A computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the steps of the intelligent acquirer contract management method of any of claims 1 to 10.
14. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the intelligent acquirement contract management method of any of claims 1 to 10.
CN202311616536.1A 2023-11-29 2023-11-29 Intelligent receipt contract management method, device and system Pending CN117635045A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311616536.1A CN117635045A (en) 2023-11-29 2023-11-29 Intelligent receipt contract management method, device and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311616536.1A CN117635045A (en) 2023-11-29 2023-11-29 Intelligent receipt contract management method, device and system

Publications (1)

Publication Number Publication Date
CN117635045A true CN117635045A (en) 2024-03-01

Family

ID=90028212

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311616536.1A Pending CN117635045A (en) 2023-11-29 2023-11-29 Intelligent receipt contract management method, device and system

Country Status (1)

Country Link
CN (1) CN117635045A (en)

Similar Documents

Publication Publication Date Title
CN108256074B (en) Verification processing method and device, electronic equipment and storage medium
CN112163681B (en) Equipment fault cause determining method, storage medium and electronic equipment
CN105654945A (en) Training method of language model, apparatus and equipment thereof
CN107229627B (en) Text processing method and device and computing equipment
US20160275444A1 (en) Procurement System
CN110597844B (en) Unified access method for heterogeneous database data and related equipment
CN110909126A (en) Information query method and device
CN115641092A (en) Method and system for realizing automatic generation of logic check by importing data check plan
WO2023007791A1 (en) Program code automatic generation system
CN117648093A (en) RPA flow automatic generation method based on large model and self-customized demand template
CN111831624A (en) Data table creating method and device, computer equipment and storage medium
CN117113947B (en) Form filling system, method, electronic equipment and storage medium
CN106570095A (en) XML data operation method and device
CN117555519A (en) SysML software demand modeling method and system based on natural language processing
CN117635045A (en) Intelligent receipt contract management method, device and system
CN117151098A (en) Relation extraction method and device and electronic equipment
CN116340352A (en) Data query method and device, storage medium and electronic equipment
CN116361316A (en) Semantic engine adaptation method, device, equipment and storage medium
CN113741864B (en) Automatic semantic service interface design method and system based on natural language processing
CN115617841A (en) Method, system, equipment and storage medium for generating data query statement
CN105630761B (en) Formula processing method and device
CN114168720A (en) Natural language data query method and storage device based on deep learning
CN117473983B (en) Unknown word collection method and device based on fuzzy matching and mutual information
CN118506783B (en) Agent-based voice interaction method and related device
CN118333157B (en) Domain word vector construction method and system for HAZOP knowledge graph analysis

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination