A contract revision method, system, device and storage medium

Through automated collection and intelligent merger of revision opinions, precise matching and implementation of the revision of contract terms, the problems of low degree of automation and low management of revision opinions in the existing technology have been solved, and efficient automation and accuracy of contract revision have been achieved. , ensuring the legality and compliance of the contract.

CN118780251BActive Publication Date: 2024-11-22STATE GRID MATERIAL CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411245384.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-11-22
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

In the field of contract management, especially when large-scale contract revisions, existing technology relies on manual review and manual modification, which leads to time-consuming and labor-intensive and easy introduction of human errors, low degree of automation, and lack of an effective revision opinion management mechanism, resulting in repairs The editorial opinions are confusing, which affects decision-making efficiency, especially in complex contracts and professional fields such as the power industry, and it is difficult to accurately match the revised opinions and contract terms.

Method used

Provide a contract revision method, by obtaining the revision opinions of multiple contract texts, merge and review based on selected revision elements, analyzing and matching the associated contract terms, realizing the revised contract The text adopts rules such as content merger, revision opinion priority merger and comprehensive merger, combined with corpus data and syntactic dependency analysis, structured contract terms, accurate matching and batch revision.

Benefits of technology

It has realized efficient automation of the contract revision process, improved the accuracy and speed of revision, optimized the management process of revision opinions, ensured the legality and compliance of contract content, and improved the overall efficiency of contract management and revision work.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118780251B_ABST
    Figure CN118780251B_ABST
Patent Text Reader

Abstract

The present application provides a contract revision method, system, device and storage medium, which is applied in the field of data processing technology. Obtain multiple revision opinions on multiple contract texts; in response to selecting at least one revision element of the multiple revision opinions, merge the multiple revision opinions based on the selected elements according to preset merging rules to obtain merged revision opinions; parse the approved revision opinions in multiple contract texts to match the contract terms associated with each approved revision opinion; based on each approved revision opinion, revise the contract terms associated with each approved revision opinion to obtain the revised contract text. Through automated collection, intelligent merging and matching and implementation of the revision of contract terms, efficient automation of the contract revision process is achieved, the accuracy and speed of revision are improved, and the management process of revision opinions is optimized, thereby improving the overall efficiency of contract management and revision work.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a contract revision method, system, device and storage medium. Background Art

[0002] In the field of contract management, traditional technologies and processes often bring many inconveniences, especially when it comes to large-scale contract revisions. When dealing with contract revisions, related technologies usually rely on manual review and manual modification, which is not only time-consuming and labor-intensive, but also prone to human errors. Even if some automated tools such as entity recognition and relationship extraction technology are used, these tools are still limited in accuracy and efficiency when faced with complex contract terms and industry-specific terminology.

[0003] In addition, when faced with revision opinions involving multiple levels and multiple roles, the existing technology lacks an effective management mechanism, which makes the processing of revision opinions chaotic and disorderly, affecting decision-making efficiency. Especially for professional fields such as the power industry, the structure of the contract text is complex, involving a large number of professional terms and specific formats. Different revision opinions may target the same contract terms, and the same revision opinion may target contract terms in multiple contracts. Therefore, it is difficult to accurately match the revision opinions with the contract terms during the revision process, making it difficult to carry out the revision work efficiently.

[0004] In summary, the limitations of existing technologies in dealing with contract revisions, including low automation and inefficient management of revision opinions, urgently need to be overcome with more advanced solutions. Summary of the invention

[0005] In view of this, the present application provides a contract revision method, system, device and storage medium, aiming to improve the automation level of contract revision and optimize the management process of revision opinions.

[0006] In a first aspect, the present application provides a contract revision method, the method comprising:

[0007] Obtain multiple revision opinions on multiple contract texts;

[0008] In response to selecting at least one revision element of the plurality of revision opinions, the plurality of revision opinions are merged according to a preset merging rule based on the selected revision element to obtain a merged revision opinion, so as to review the merged revision opinion to obtain a revision opinion that has been reviewed and approved; the revision elements include reason category, contract text category, contract text content, content before revision, and content after revision; the preset merging rule includes one or more of content merging, revision opinion priority merging, and comprehensive merging;

[0009] Parsing the approved revision opinions, and matching the contract clauses associated with the approved revision opinions in multiple contract texts;

[0010] Based on each approved revision opinion, the contract clauses associated with each approved revision opinion are revised to obtain a revised contract text.

[0011] Optionally, when the preset merging rule is content merging, merging the multiple revision opinions according to the preset merging rule to obtain merged revision opinions includes:

[0012] Extracting entities from the revised contents of the plurality of revision opinions;

[0013] Based on the merging priority corresponding to each entity in the revised content of the multiple revision opinions, the multiple revision opinions are merged to obtain merged revision opinions. The merging priority is used to determine the merging method for merging similar revision opinions. The similar revision opinions are calculated based on the similarity of the multiple revision opinions.

[0014] Optionally, the parsing of the approved revision opinions in multiple contract texts to match the contract clauses associated with each approved revision opinion includes:

[0015] Obtain corpus data and contract terms related to contract business scenarios;

[0016] Establishing a mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms;

[0017] Structuring the contract terms in the plurality of contract texts based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms;

[0018] The approved revision opinions are parsed, and the contract terms associated with the approved revision opinions are matched in the structured contract terms.

[0019] Optionally, the step of structuring the contract terms in the plurality of contract texts based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms includes:

[0020] Extracting metadata from the multiple contract texts to generate a set of contract clauses;

[0021] Performing syntactic dependency analysis on each contract clause in the contract clause set to obtain a syntactic dependency tree corresponding to each contract clause;

[0022] Based on the syntactic dependency tree corresponding to each contract clause, the contract clauses in multiple contract texts are structured using the mapping relationship between the corpus data corresponding to the contract business scenario and the contract clauses to obtain structured contract clauses.

[0023] Optionally, the obtaining of corpus data and contract terms related to the contract business scenario includes:

[0024] Build the contract text management library, procurement type management library, material type management library, contract clause management library and attribute tag management library corresponding to the business scenarios;

[0025] Based on the contract text management library, procurement type management library, material type management library, contract clause management library and attribute label management library corresponding to the business scenario, corpus data and contract clauses related to the business scenario are set respectively.

[0026] Optionally, the parsing of the approved revision opinions in multiple contract texts to match the contract clauses associated with each approved revision opinion includes:

[0027] Pre-set the keywords in the contract terms and the corresponding levels of the keywords;

[0028] By parsing the approved revision opinions, keywords are extracted from the pre-revised content and the post-revised content of each approved revision opinion;

[0029] Determine the similarity with the contract clauses in the multiple contract texts according to the keywords corresponding to each revision opinion and the levels corresponding to the keywords;

[0030] The contract clauses associated with each approved revision opinion are matched according to the similarity in multiple contract texts, and the revision opinions are displayed in descending order according to the levels corresponding to the keywords in the contract clauses.

[0031] Optionally, the step of revising the contract clauses associated with each approved revision opinion based on each approved revision opinion to obtain a revised contract text includes:

[0032] Locate the contract clauses associated with each approved revision opinion in the contract text based on each approved revision opinion;

[0033] According to the revised content of the revision opinions, the contract clauses associated with each approved revision opinion are revised in batches to obtain multiple revised contract texts;

[0034] A revision report is generated based on the contract text before revision and the corresponding contract text after revision; the revision report at least includes revision information and a revision difference comparison table.

[0035] In a second aspect, the present application provides a contract revision system, the system comprising the contract revision opinion collection module, the revision opinion business review module and the revision opinion confirmation module;

[0036] The contract revision opinion collection module is used to obtain multiple revision opinions on multiple contract texts;

[0037] A revision opinion business review module is used for, in response to selecting at least one revision element of the plurality of revision opinions, merging the plurality of revision opinions according to a preset merging rule based on the selected revision element to obtain a merged revision opinion, so as to review the merged revision opinion to obtain a revision opinion that has been reviewed and approved; the revision elements include reason category, contract text category, contract text content, content before revision and content after revision; the preset merging rule includes one or more of content merging, revision opinion priority merging and comprehensive merging;

[0038] The revision opinion confirmation module is used to parse the approved revision opinions in multiple contract texts to match the contract terms associated with each approved revision opinion; based on each approved revision opinion, the contract terms associated with each approved revision opinion are revised to obtain the revised contract text.

[0039] In a third aspect, the present application provides a device comprising a memory and a processor, wherein the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes, so that the device executes the contract revision method described in any one of the first aspects above.

[0040] In a fourth aspect, the present application provides a computer storage medium having codes stored therein. When the codes are executed, a device executing the codes implements the contract revision method described in any one of the first aspects above.

[0041] The present application provides a contract revision method, system, device and storage medium, which are applied in the field of data processing technology. When executing the method, first obtain multiple revision opinions on multiple contract texts; then in response to selecting at least one revision element of the multiple revision opinions, merge the multiple revision opinions based on the selected revision elements according to the preset merging rules to obtain the merged revision opinions, so as to review the merged revision opinions to obtain the revised revision opinions that have been reviewed and approved; the revision elements include reason categories, contract text categories, contract text content, content before revision and content after revision; the preset merging rules include one or more of content merging, revision opinion priority merging and comprehensive merging; then parse the revised revision opinions that have been reviewed and approved, and match the contract clauses associated with the revised revision opinions in multiple contract texts; finally, based on each revised revision opinion that has been reviewed and approved, revise the contract clauses associated with each revised revision opinion that has been reviewed and approved to obtain the revised contract text. In this way, by automatically collecting, intelligently merging and reviewing revision opinions, and accurately matching and implementing revisions to contract terms, this method achieves efficient automation of the contract revision process, significantly improves the accuracy and speed of revisions, and at the same time optimizes the management process of revision opinions, ensuring the legality and compliance of contract content, thereby greatly improving the overall efficiency of contract management and revision work. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A pipeline entity relationship extraction method of a related technology provided in an embodiment of the present application;

[0044] Figure 2A A schematic diagram of the structure of a contract revision system provided in an embodiment of the present application;

[0045] Figure 2B A schematic diagram of a contract text template management unit viewing a standard contract template provided in an embodiment of the present application;

[0046] Figure 3 A method flow chart of a contract revision method provided in an embodiment of the present application;

[0047] Figure 4 A flowchart of a possible implementation method of step S2 provided in an embodiment of the present application;

[0048] Figure 5A schematic diagram of a revision opinion query provided in an embodiment of the present application;

[0049] Figure 6 A flowchart of a possible implementation method of step S3 provided in an embodiment of the present application;

[0050] Figure 7 A schematic diagram of the structure of a material type management library provided in an embodiment of the present application;

[0051] Figure 8 A flowchart of a possible implementation method of step S33 provided in an embodiment of the present application;

[0052] Fig. 9 A schematic diagram of a structured contract clause provided for an embodiment of the present application;

[0053] Fig.10 A schematic diagram of a modification interface provided in an embodiment of the present application;

[0054] Fig.11 A flowchart of a possible implementation method of step S3 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] As mentioned above, the relevant technologies usually rely on manual review and modification when dealing with contract revision. Even if some automated tools are used, such as entity recognition and relationship extraction technology, the accuracy and efficiency of these tools are still limited when faced with complex contract terms and industry-specific terminology. Figure 1 A pipeline entity relationship extraction method of a related technology provided in an embodiment of the present application. Figure 1 As shown in the figure, the traditional pipeline entity relationship extraction method decomposes the task into two steps: named entity recognition and relationship extraction. When the entity relationship triples are extracted based on unstructured text, although the process is clear and easy to understand, its step-by-step processing characteristics easily lead to error accumulation, affecting the accuracy of the final result. The current contract text structuring technology uses natural language processing to realize the automatic classification of contract terms and the extraction of key information, which significantly improves the efficiency of contract management. However, these technologies face challenges in accuracy and applicability when dealing with specific business scenarios such as material procurement contracts in the power industry.

[0056] In view of this, the present application provides a contract revision method, system, device and storage medium. When executing the method, first obtain multiple revision opinions on multiple contract texts; then in response to selecting at least one revision element of the multiple revision opinions, merge the multiple revision opinions based on the selected element according to the preset merging rule to obtain the merged revision opinions, so as to review the merged revision opinions to obtain the approved revision opinions; the revision elements include reason categories, contract text categories, contract text content, content before revision and content after revision; the preset merging rules include one or more of content merging, revision opinion priority merging and comprehensive merging; then parse the approved revision opinions in multiple contract texts to match the contract clauses associated with each approved revision opinion; finally, based on each approved revision opinion, revise the contract clauses associated with each approved revision opinion to obtain the revised contract text.

[0057] In this way, the above scheme can systematically process and integrate revision opinions from different sources, reduce duplication of work, and improve revision efficiency and accuracy. It solves the problems of low automation and low efficiency in the management of revision opinions in the prior art, especially when dealing with complex contracts, such as material procurement contracts in the power industry, which can significantly improve the quality and speed of revision work. In addition, through intelligent matching and automated revision, this application also enhances the transparency and controllability of the contract revision process, which helps to maintain the legal effect and commercial value of the contract.

[0058] Figure 2A A schematic diagram of the structure of a contract revision system provided in an embodiment of the present application. Figure 2A As shown, the contract revision system 1 at least includes a contract revision opinion collection module 11 , a revision opinion business review module 12 and a revision opinion confirmation module 13 .

[0059] The contract revision opinion collection module 11 is used to obtain multiple revision opinions on multiple contract texts. In this embodiment, a customer who needs to provide revision opinions can upload or record the revision opinions to the contract revision opinion collection module 11 by logging into the contract revision system 1 .

[0060] The revision opinion business review module 12 is used to respond to the selection of at least one revision element of multiple revision opinions, merge the multiple revision opinions according to the selected element according to the preset merging rules to obtain merged revision opinions, so as to review the merged revision opinions to obtain revised opinions that have been reviewed and approved.

[0061] As mentioned above, when dealing with the solicitation of opinions on the revision of various types of contract texts, we will encounter a large number of diverse revision opinions from staff in different units and positions. These opinions may involve different parts of the contract, from payment terms, breach of contract liability to confidentiality agreements, and the focus and priority of each opinion may also be different. In such a case, the traditional method is often to list all the opinions one by one and require the editors to read and evaluate them one by one. This method is not only time-consuming, but also prone to missing or confusing information, affecting the quality of decision-making.

[0062] In this embodiment, by automatically classifying and integrating the revision opinions, similar or related opinions are grouped together to form a more systematic summary of opinions. For example, all revision opinions on payment terms will be grouped together, and all revision opinions on confidentiality clauses will be sorted separately according to the content of the opinions corresponding to the payment terms. By analyzing and integrating the revision opinions involving the same type of clauses, commonalities and differences are found, and the same or similar contract clauses are merged into one revision opinion based on the preset merging rules, so that the editors can more efficiently review and compare different opinions on the same topic, so as to make more accurate decisions. Among them, the revision elements include reason categories, contract text categories, contract text content, content before revision and content after revision, etc.; the preset merging rules include one or more of content merging, revision opinion priority merging and comprehensive merging.

[0063] Through such modular display and screening functions, the reviewer of the revision opinion can manage the revision work in a more systematic and orderly manner, ensuring that each revision opinion can be properly noticed and handled, thereby improving the scientificity and efficiency of the revision work. This systematic approach also helps to reduce omissions and duplication of work in the revision process, and promotes the transparency and fairness of contract revision decisions.

[0064] The revision opinion confirmation module 13 is used to parse the approved revision opinions in multiple contract texts and match the contract terms associated with each approved revision opinion; based on each approved revision opinion, the contract terms associated with each approved revision opinion are revised to obtain the revised contract text.

[0065] In an optional embodiment, the contract revision system 1 provided in the embodiment of the present application may also include a contract standard text library management module 2, which is used to manage and maintain the standard contract text library of the enterprise to ensure the standardization, accuracy and efficiency of the contract text. Specifically, the contract standard text library management module 2 includes a contract text template management unit, a contract text information query unit and a contract text library analysis unit.

[0066] Among them, the contract text template management unit is responsible for the creation, editing, storage and viewing of various standard contract templates. For example, the contract text template management unit can store various standard contract templates commonly used by enterprises, such as sales contracts, purchase contracts, lease contracts, etc. Figure 2B A schematic diagram of a contract text template management unit viewing a standard contract template provided in an embodiment of the present application. It should be noted that the contract standard text library management module 2 supports the storage of contract texts in multiple formats, such as Word, PDF, etc., to meet the needs of different departments and scenarios. In addition, the contract standard text library management module 2 can also regularly review and update the contract template to ensure that the template content complies with changes in laws, regulations and corporate policies.

[0067] The contract text information query unit provides a variety of query methods to help users quickly locate the required contract text, and supports full-text search, conditional filtering, and classified browsing. Specifically, users can enter keywords or phrases, and the system will perform a full-text search in the contract text library and return matching results. The contract text information query unit also supports conditional filtering, that is, it supports filtering by contract type, signing date, contract status (such as pending, signed, invalid), and other conditions. In an optional embodiment, the contract text information query unit can organize the contract text according to a preset classification system, and users can search for contracts by browsing the classification tree. For the queried contract text, a detailed information viewing function is provided, including contract content, attachments, related party information, etc.

[0068] The contract text library analysis unit uses data analysis technology to conduct in-depth mining and analysis of the data in the contract text library to provide decision support for enterprises. Specifically, it can conduct statistical analysis on basic information such as the number, type, and signing date of contracts, and generate reports and charts. In addition, it can analyze the trends and rules of contract signing based on historical data and predict future contract needs. And it can use natural language processing (NLP) technology to perform semantic analysis on contract texts to identify potential contract risks and non-compliant clauses. Based on the user's historical behavior and data analysis results, it recommends appropriate contract templates or contract clauses to users.

[0069] In an optional embodiment, the contract revision system 1 provided in the embodiment of the present application also includes a contract text relationship management module 3, which is mainly used to manage and maintain the relationship between contract corpus, contract terms, and contract text to ensure the integrity and consistency of the contract system. Specifically, it may include a contract text management unit, a procurement type management unit, a material type management unit, a contract term management unit, and an attribute label management unit, which are used to define the relationship between contract corpus, contract terms, and contract text, such as parent-child relationship, reference relationship, etc. In each unit of the contract text relationship management module 3, dynamic adjustment and maintenance of the relationship are supported to adapt to the needs of changes in the enterprise business.

[0070] It should be noted that the embodiments of the present application do not limit the execution subject of the contract revision method. For example, the contract revision method of the embodiments of the present application can be applied to data processing devices such as terminal devices or servers. The terminal device can be a smart phone, a computer, a personal digital assistant (PDA) or a tablet computer. The server can be an independent server, a cluster server or a cloud server.

[0071] In order to more clearly understand the embodiments below, the embodiments of the present application explain structured text and unstructured text.

[0072] Structured text refers to data organized into text with a fixed format and clear fields, usually corresponding to a table, database or specific data structure. In the contract environment, this means that the various parts of the contract, such as titles, clauses, signing dates, information about both parties, etc., are clearly labeled and classified. The characteristics of structured text include: Fixed format: Data is presented in the form of rows and columns, similar to a spreadsheet or database table. Easy to process: Due to the clear fields, data can be easily searched, sorted and analyzed through query languages ​​or data processing software. Standardization: The structure of the data is predefined, which facilitates automated processing and data exchange between systems. For example, in a lease contract, the information is broken down into: lease contract number; names of the lessor and lessee; lease start and end dates; monthly rent amount; and an overview of the breach clause. This information will be clearly marked in the structured text, such as "Lease Contract Number: CT12345", so that the system can quickly identify and process it.

[0073] Unstructured text refers to text that has no fixed format or predefined structure and usually contains natural language, such as the complete text of an article, email, social media post, or long contract. In the text of a contract, unstructured text refers to the detailed description of the terms of the contract, which contains various legal terms and conditions without clear field distinctions. The characteristics of unstructured text include: Free format: The text can be of any length and has no fixed fields or tags. Difficult to process: Due to the lack of structure, more complex technologies such as natural language processing (NLP) are required to analyze and extract specific information. Information intensive: Although data is rich, it may be time-consuming and laborious to extract and parse the required information. For example, a complete contract text contains a detailed description of the terms, such as: Party A agrees to pay Party B a total advance payment of RMB 5,000 within thirty days from the date of the effective date of the contract, and Party B shall not enter into similar agreements with other competitors during this period unless Party A has given prior written approval. This text has no clear fields, but uses natural language to describe the terms of the contract, which makes it more difficult to automatically extract key information from the text.

[0074] In contract management, converting unstructured text into structured data is a key step that helps the system more effectively manage the contract lifecycle, including contract creation, review, execution, and monitoring. By using natural language processing techniques such as entity recognition, keyword extraction, and semantic analysis, unstructured text can be parsed to extract key information such as contract dates, amounts, terms, etc., and then stored in a structured format for subsequent retrieval and analysis. This conversion is critical to automating the contract management process, improving efficiency, and reducing errors.

[0075] In order to be able to understand the characteristics and technical contents of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The attached drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, in order to simplify the drawings, well-known structures and devices can be simplified for display. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0076] Figure 3 A method flow chart of a contract revision method provided in an embodiment of the present application. Figure 3 As shown, the contract revision method provided in the embodiment of the present application may include:

[0077] S1, obtain multiple revision opinions on multiple contract texts.

[0078] In this embodiment, the revision opinions for multiple contract texts from different sources (such as legal departments, business departments, customer feedback, etc.) are collected. These opinions may involve various types such as modification, addition, and deletion of contract terms. Specifically, an online feedback system can be established through the contract revision system to centrally collect these opinions to ensure that each opinion has a clear source and points to a specific contract text and clause. The user logs in to the contract revision system and fills in the revision opinions through the contract opinion collection module therein. Since the user uploads the revision opinions by logging in to the contract revision system, the revision opinions can automatically carry the user's identification and other information related to the user information.

[0079] In an optional embodiment, Table 1 is a possible editing opinion header provided in an embodiment of the present application.

[0080] Table 1

[0081]

[0082] As shown in Table 1, users can provide revision suggestions by referring to the above table.

[0083] S2, in response to selecting at least one revision element of multiple revision opinions, merging the multiple revision opinions based on the selected revision element according to a preset merging rule to obtain a merged revision opinion, so as to review the merged revision opinions to obtain revised opinions that have been reviewed and approved.

[0084] In this embodiment, the revision elements include reason category, contract text category, contract text content, content before revision and content after revision. Among them, the reason category refers to the reason for proposing a revision opinion, such as legal compliance, changes in business requirements, customer requirements, etc. The contract text category refers to the type of contract involved, such as a purchase contract, a sales contract, a service contract, etc. The contract text content refers to specific contract clauses or paragraphs. Content before revision: refers to the original contract text that needs to be revised. Content after revision: refers to the content of the revised contract text.

[0085] In this embodiment, multiple revision opinions are sorted, classified and merged according to the above-mentioned revision elements. The reviewer of the revision opinion can select one of the following elements separately to merge and view the revision opinions, or set several parallel elements for integrated viewing, so as to facilitate the reviewer of the revision opinion to modularly review the revision opinions and improve the scientificity and orderliness of the revision work. Specifically, for the same or similar revision opinions of the same contract text or similar contract clauses, they can be merged into one merged revision opinion to reduce the complexity and redundancy of subsequent processing.

[0086] In this embodiment, a set of merging rules is pre-set before merging. These rules can be based on different factors, such as the proposer of the revision opinions, importance, consistency, etc. For example, the rules may stipulate that when multiple revision opinions are for the same contract clause and the revision suggestions are the same, they should be directly merged into one revision opinion; or when the revision suggestions are slightly different but the core meaning is the same, these opinions should be combined to form a more comprehensive merged revision opinion.

[0087] Specifically, the preset merging rules may include content merging, revision opinion priority merging, and comprehensive merging. Content merging means that if multiple revision opinions are aimed at the same contract clause and the modification suggestions are the same or similar, these opinions can be directly merged into one revision opinion. Priority merging means that if there are priority differences between multiple revision opinions (such as some opinions come from higher-level decision makers or more critical stakeholders), they can be merged based on priority, retaining opinions with higher priorities. Comprehensive merging means that in some cases, the contents of multiple revision opinions are integrated to form a more comprehensive and accurate merged revision opinion.

[0088] The merged revision opinions need to be reviewed by professionals to ensure that they comply with laws and regulations, business needs and the spirit of the contract. During the review process, it may be discovered that the revision opinions need to be further adjusted or rejected. Specifically, in this embodiment, an audit process is established to clarify the audit standards and responsible persons. A workflow management system can be used to track the progress of the audit and record the audit opinions and results. Since in this embodiment, the auditor can automatically merge and display multiple revision opinions according to the above method, the auditor can view the revision opinions more clearly and efficiently, ensuring the accuracy and efficiency of the revision process.

[0089] In an optional embodiment, for the adoption or rejection of the revision opinions, the adoption results will be fed back to the corresponding party that proposed the revision opinions, ensuring a closed-loop revision workflow.

[0090] In an alternative embodiment, Figure 4 A possible implementation flow chart of step S2 provided in the embodiment of the present application, combined with Figure 4 As shown, when the preset merging rule is content merging, merging the multiple revision opinions according to the preset merging rule to obtain the merged revision opinions includes:

[0091] S21, extracting entities from the revised contents of the plurality of revision opinions.

[0092] In this embodiment, key entities are extracted from the revised contents of multiple revision opinions. These entities are usually specific objects or contents mentioned in the revision opinions, such as contract names, clause numbers, specific texts before and after revision, etc. The purpose of entity extraction is to identify the core information in the revision opinions for subsequent merging. Entity extraction can be implemented using the named entity recognition (NER) method in natural language processing (NLP) technology. Through the trained model, key entities in the text can be automatically identified and used as the basis for merging.

[0093] S22, merging the multiple revision opinions based on the merging priorities corresponding to the entities in the revision contents of the multiple revision opinions to obtain merged revision opinions.

[0094] The merge priority is used to determine the merging method for merging similar revision opinions. Similar revision opinions are calculated based on the similarity of multiple revision opinions.

[0095] In this embodiment, after extracting the entities, the system calculates the similarity between different revisions. This is achieved by comparing the text content of the revisions, the entities involved, the revised terms, etc. The similarity calculation is the basis for merging, which helps the system identify which revisions can be considered similar or related.

[0096] Based on the similarity calculation, the system assigns a merge priority to each entity or revision opinion. This priority can be based on factors such as the source of the revision opinion (such as senior management's opinions have a higher priority), the urgency of the revision, and the consensus of the revision opinions. The merge priority is used to determine which opinions should be given priority when merging similar revision opinions. It should be noted that the "merge priority" here does not directly determine which revision opinion will be fully adopted, but is used to guide how to merge similar but different opinions.

[0097] Specifically, after entity extraction, similarity calculation and merging priority, the system can start merging revision opinions. Similar revision opinions will be merged according to their priority, and opinions with high priority will dominate the merged revision opinions. The merged revision opinions will contain the core points of all relevant opinions while avoiding duplication and conflict.

[0098] For example, for an entity such as "term", if different values ​​or conditions are provided in different revision opinions for the same contract clause, these different opinions can be merged into one item, but the respective values ​​are retained in the merged revision opinions, and other identical contents are completely merged, so that the reviewer can only conduct a comparative review of the modified parts in one revision opinion. It should be noted that during the merging process, it should be ensured that the original revision opinions corresponding to each merged opinion can be traced back. This can be achieved by adding reference links or annotations to the merged opinions. For example, based on different numerical annotation sources, the original revision opinions can be further viewed by clicking on a specific value.

[0099] In this embodiment, through the structured contract text revision method, specific revision opinions and suggestions can be integrated and classified, and the revision opinions involving various clauses can be analyzed and integrated, which is convenient for revision personnel to review and audit, and comprehensively consider the revision opinions, thereby improving the efficiency of revision opinion analysis.

[0100] It should be noted that the above-mentioned contract opinion merging method can be applied to the revision opinion query process. Figure 5 A schematic diagram of a revision opinion query provided in an embodiment of the present application. Figure 5 As shown, the contents before modification can be merged, similar or identical clauses corresponding to different contract text numbers can be displayed together, and the number of contracts involved can be further displayed.

[0101] S3, parse the approved revision opinions, and match the contract clauses associated with the approved revision opinions in multiple contract texts.

[0102] In an alternative embodiment, Figure 6 A possible implementation flow chart of step S3 provided in the embodiment of the present application, combined with Figure 6 As shown, the matching of the approved revision opinions in multiple contract texts and the contract clauses associated with each approved revision opinion include:

[0103] S31, obtaining corpus data and contract terms related to the contract business scenario.

[0104] In this step, corpus data closely related to the contract business scenario is collected. These data may include industry terms, common expressions, legal provisions, etc. At the same time, it is also necessary to obtain the specific terms and conditions in the contract text to be revised. For contracts involved in specific business scenarios, such as the material procurement contract text in the power industry, the materials involve 38 major categories, 392 medium categories, 5089 minor categories and 22 voltage levels. The contracts used to purchase different materials are different. Traditional general corpus cannot contain knowledge of specific business scenarios. This embodiment can increase the accuracy of data processing in the contract revision process of specific scenarios by obtaining corpus data and contract terms related to the contract business scenario.

[0105] In an optional embodiment, corpus data and contract terms related to the contract business scenario are obtained, including: constructing a contract text management library, a procurement type management library, a material type management library, a contract term management library, and an attribute label management library corresponding to the business scenario; and respectively setting the corpus data and contract terms related to the business scenario based on the contract text management library, the procurement type management library, the material type management library, the contract term management library, and the attribute label management library corresponding to the business scenario.

[0106] In this embodiment, in the process of acquiring corpus data and contract terms related to the contract business scenario, a series of management libraries are constructed to organize, classify and retrieve a large amount of information closely related to the contract business scenario.

[0107] Specifically, first, it is necessary to build a contract text management library, a procurement type management library, a material type management library, a contract terms management library, and an attribute label management library. These management libraries are used to store and manage different types of contract texts, procurement types, material types, contract terms, and attribute labels. After building these management libraries, the next step is to set up and organize corpus data and contract terms related to business scenarios based on these libraries. This usually involves extracting information from existing data sources such as contract texts, procurement records, and bills of materials, and storing them in corresponding management libraries. In the process of setting up corpus data and contract terms, it is also necessary to establish the association relationship between them and create indexes for them. In this way, when you need to query or retrieve a specific contract term or data, you can quickly locate the relevant information.

[0108] The contract text management library is used to store and manage various types of contract texts, including historical contracts, standard contracts, revised contracts, etc. Contract texts can be classified and stored according to factors such as contract type (such as procurement contracts, sales contracts, lease contracts, etc.), industry characteristics, signing time, etc. Indexes are created for contract texts for quick retrieval and access.

[0109] The procurement type management library is used to define and manage different procurement types, which are usually associated with specific contract terms and business processes. Based on business needs and market practices, a series of procurement types are defined, such as goods procurement, service procurement, engineering procurement, etc. Common contract terms and templates are associated with each procurement type for quick reference when creating a new contract.

[0110] The material type management library is used to define and manage the material types involved in the contract, which helps to accurately describe the material characteristics and requirements in the contract terms. According to business needs and market practices, define a series of material types, such as raw materials, parts, finished products, etc. Associate attribute tags for each material type, such as specifications, models, brands, quality standards, etc., so that material requirements can be described in detail in the contract terms. Figure 7 A schematic diagram of the structure of a material type management library provided in an embodiment of the present application. In the power scenario, users can add or delete relevant materials of a specific business scenario in the material type management library according to their needs.

[0111] The contract clause management library is used to store and manage various contract clause templates and key clause information for quick reference and modification when creating a new contract. Contract clause templates are extracted from existing contract texts and sorted and optimized according to business needs. Contract clauses are classified and stored according to clause types (such as performance clauses, liability for breach of contract, dispute resolution, etc.) and associated business scenarios. Contract clauses are indexed for quick retrieval and reference.

[0112] The attribute tag management library is used to define and manage attribute tags related to contract business scenarios. These tags are used to mark and classify contract texts, materials, procurement types, etc. Based on business needs and data characteristics, a series of attribute tags are defined, such as contract type, signing date, material attributes, procurement type, etc. Corresponding attribute tags are associated with each data item (such as contract text, material records, etc.) to facilitate multi-dimensional classification and retrieval.

[0113] The above databases can regularly update and maintain the attribute tag management library as the business develops and the amount of data increases.

[0114] S32, establishing a mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms.

[0115] In this embodiment, the above-mentioned related technologies usually use entity recognition methods to mark key elements in the contract and extract important elements in the contract, but the errors in the structuring process of specific contract texts are large, the quality is not high, and specific industries cannot be structured.

[0116] In this embodiment, a mapping relationship model between different materials and contract terms is established to assist users in quickly and specifically proposing revision opinions for different materials applicable to different procurement contracts. Specifically, based on the corpus data and contract terms corresponding to the contract business scenario, a mapping relationship between the corpus data and contract terms corresponding to the contract business scenario is established. This mapping relationship helps to convert unstructured contract terms into a structured form that is easier to understand and process, and also provides a basis for subsequent matching of revision opinions.

[0117] S33, based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms, the contract terms in the multiple contract texts are structured to obtain structured contract terms.

[0118] In this embodiment, by using the established mapping relationship, the system can perform structured processing on the contract clauses in multiple contract texts. This means converting the contract texts into structured data, in which each clause is classified and labeled, so as to facilitate computer understanding and processing, and facilitate the merging of revision opinions and subsequent matching. This process may involve natural language processing techniques, such as semantic analysis and entity recognition, to ensure accurate classification of clauses.

[0119] Specifically, contract clause structuring is to extract structured information from contract clauses in a contract instance and generate the structure of the full text contract through rule mapping. To obtain structured data of contract clauses, we can perform syntactic analysis on the contract clauses, extract the dependency between words in the sentence based on the contract clauses, and use the type of the core word in the sentence to determine the type of contract rules.

[0120] In order to describe the structured and normalized contract rules, we first formally define the rule metamodel based on the contract metamodel, define the heuristic rules in the contract clauses, then perform dependency syntax analysis on the clauses of the contract instance to generate a dependency tree, and finally map it to the clause structured model.

[0121] In an alternative embodiment, Figure 8 A possible implementation flow chart of step S33 provided in the embodiment of the present application, combined with Figure 8 As shown, the contract terms in multiple contract texts are structured based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms, which may include:

[0122] S331, extracting metadata from multiple contract texts to generate a set of contract clauses.

[0123] In this embodiment, in order to classify the clauses in the contract text according to certain rules or standards, a contract clause rule classification is established, which can be based on the function, nature, legal significance, etc. of the contract clauses. The purpose of classification is to facilitate subsequent processing so that the same or similar clauses can be processed uniformly.

[0124] Specifically, the purpose of this step is to separate the basic contract information from the contract terms and generate a contract terms set based on the contract terms. By extracting the contract metadata, the basic information of the contract in the contract instance is obtained, including the contract name, number, parties, location and other key information, and this information is separated from the contract terms. Furthermore, in order to obtain the contract terms information, a contract terms dictionary is established, and each term is separated through dictionary matching, and finally a contract terms set is obtained.

[0125] Fig. 9 A schematic diagram of a structured contract clause provided in an embodiment of the present application. Fig. 9 As shown, metadata is imported based on the contract instance and classified according to the contract clause rules, such as basic information, time, clauses, etc., to divide the metadata into basic contract information and contract clause sets.

[0126] S332, performing syntactic dependency analysis on each contract clause in the contract clause set to obtain a syntactic dependency tree corresponding to each contract clause.

[0127] Syntactic dependency analysis is a natural language processing technology used to analyze the dependency relationship between words in a sentence. It usually reveals the tree structure of the sentence's syntactic structure and shows the dependency relationship between grammatical components such as subject, predicate, object, attributive, adverbial, and complement in the sentence. In the process of structuring contract terms, syntactic dependency analysis can help understand the semantic relationship between the various components within the contract terms. By constructing a syntactic dependency tree, the dependency and modification relationship between the various words or phrases in the contract terms can be clearly seen.

[0128] See also Fig. 9 As shown in the figure, we take a single clause as a unit and perform syntactic analysis on each single sentence of the clause. We first mark the contract clauses, such as Chinese word segmentation and part-of-speech tagging, and then perform dependency syntactic analysis. We extract dependency tuples to identify the core words in the sentence and build a syntactic dependency tree. Dependency syntactic analysis is the basis for structuring the contract clauses below and provides structured data for the generation of smart contracts.

[0129] S333, based on the syntactic dependency tree corresponding to each contract clause, the contract clauses in the multiple contract texts are structured using the mapping relationship between the corpus data corresponding to the contract business scenario and the contract clauses to obtain structured contract clauses.

[0130] In this embodiment, the contract terms are structured by combining the mapping relationship between the corpus data corresponding to the contract business scenario and the contract text terms, as well as the syntactic dependency tree obtained in the previous step. This usually involves extracting key information (such as subject, predicate, object, etc.) from the syntactic dependency tree, and mapping this information to a preset structured template according to the mapping relationship. In this way, the original contract terms text is converted into structured data with clear structure and semantic information.

[0131] Specifically, see Fig. 9 As shown in the figure, the contract terms can be matched and classified by constructing a rule metamodel, the type of the clause rule can be judged according to the core words corresponding to the syntactic dependency tree, and the structured contract terms can be generated through rule matching. The rule matching process includes rule clause classification and clause structuring. Based on the results of the above dependency syntactic analysis, the type of the clause rule can be judged according to the core words, and the structured contract terms can be generated through rule matching.

[0132] S34, parsing the approved revision opinions, and matching the contract terms associated with the approved revision opinions in the structured contract terms.

[0133] In this embodiment, the approved revision opinions are parsed, and the contract terms associated with each revision opinion are searched and matched in the structured contract terms. This step is a key link in the revision process, which ensures that the revision opinions can be accurately applied to the corresponding contract terms.

[0134] In this embodiment, by structuring the contract terms and establishing a mapping relationship between the corpus data and the contract terms, the content of the contract terms can be understood and analyzed more accurately.

[0135] In this embodiment, the structured contract terms convert text content that may be vague or diverse into data with a clear structure and format. This clarity and consistency helps to reduce matching errors caused by differences in textual expressions. In addition, on the basis of establishing mapping relationships and structured contract terms, matching rules can be more easily defined and applied. Since structured contract terms already have clear structure and semantic information, these rules can be executed more accurately. Moreover, for complex scenarios involving multiple related terms or requiring matching across contract texts, structured contract terms can support more complex matching logic. For example, matching can be performed based on the association relationship or semantic similarity between contract terms, rather than relying solely on simple keyword matching. This capability makes the matching results more accurate and comprehensive.

[0136] In summary, the above embodiments can achieve automatic matching between revision opinions and contract terms by structuring contract terms and defining matching rules. This not only improves the efficiency of revision work, but also reduces the need for manual intervention and potential human errors. Automatic matching can also support batch processing and rapid response to meet the growing business needs of enterprises.

[0137] In addition, the structured contract text revision method can further filter and sort the revision opinions by setting keywords or topic tags, so that the revision staff can give priority to certain key areas or high-priority opinions, and put the secondary opinions aside for later processing. This method not only improves the efficiency of the revision opinion analysis, but also ensures the systematic and comprehensive nature of the revision process, avoiding the neglect of important opinions due to information overload.

[0138] S4, based on each approved revision opinion, revise the contract clauses associated with each approved revision opinion to obtain a revised contract text.

[0139] According to the marked contract clauses and corresponding revision opinions, actual modifications are made to the contract text, which may include adding new clauses, deleting old clauses, modifying the content of existing clauses, etc. Fig.10 A schematic diagram of a modification interface provided in an embodiment of the present application. Fig.10 As shown, this embodiment batch revises the contract clauses associated with each approved revision opinion, such as 2022 revision opinions, and finally obtains the revised contract text.

[0140] In an optional embodiment, the contract clause associated with each approved revision opinion is revised based on each approved revision opinion to obtain a revised contract text, including: locating the contract clause associated with each approved revision opinion in the contract text based on each approved revision opinion; batch revising the content of the contract clause associated with each approved revision opinion according to the revision content of the revision opinion to obtain multiple revised contract texts; generating a revision report based on the contract text before revision and the corresponding contract text after revision; the revision report at least includes revision information and a revision difference comparison table.

[0141] In this embodiment, for the clauses that are determined to be revised, this method will locate the confirmed revision opinions in the contract clauses of the specific revised contract text according to the text serial number, text name, chapter serial number, and detailed serial number of the clauses to be revised. After clicking on the execution, batch restoration and replacement will be performed to generate a revised version of the contract text based on the original uploaded contract text. Specifically, it is for review by the approval and application personnel, and a revision difference comparison table can be generated to summarize which clauses and texts are specifically revised in the contract text revision for the personnel to review. At the same time, the revision of the version of the contract clauses will be analyzed, and the number and frequency of revisions of the key vocabulary clauses will be large, and which contract texts have more revised content will be analyzed in multiple dimensions.

[0142] In this embodiment, by automatically collecting, intelligently merging and reviewing revision opinions, and accurately matching and implementing revisions to contract terms, this method achieves efficient automation of the contract revision process, significantly improves the accuracy and speed of revisions, and at the same time optimizes the management process of revision opinions, ensuring the legality and compliance of contract content, thereby greatly improving the overall efficiency of contract management and revision work.

[0143] In the embodiment of the present application, the above Figure 3 There are multiple possible implementations of step S2, which are described below. It should be noted that the implementations described below are only exemplary and do not represent all implementations of the embodiments of the present application.

[0144] Fig.11 A possible implementation flow chart of step S2 provided in the embodiment of the present application. Fig.11 As shown, the embodiment of the present application provides a contract revision method, which may also include:

[0145] S25, pre-setting the keywords in the contract terms and the levels corresponding to the keywords.

[0146] In this embodiment, first, a set of keywords is defined based on business needs and the characteristics of the contract text. These keywords should cover common terms, phrases or specific expressions in contracts so that they can be matched between revision opinions and contract terms. Each keyword is assigned a level to reflect its importance or sensitivity in the contract terms (e.g., if the high-priority positive element is settlement, all revision opinions involving settlement are integrated; if the high-priority negative element is breach of contract, all elements involving breach of contract are hidden). For example, warranty terms may be considered level 1 keywords, settlement terms may be level 2, and general descriptions may be classified as lower levels.

[0147] For example, taking the power contract as an example, for the setting of key words, this method can set the key words and levels in the contract terms, such as level 1-warranty, level 2-settlement, level 3-acceptance, and compare the revision opinions containing such key words with the contract text in the full text library, and arrange them according to the degree of overlap at level 1, level 2, and level 3, so as to quickly arrange the importance of the revised terms and enable the revision reviewers to review them according to importance.

[0148] S26, extracting keywords from the pre-revised content and the post-revised content of each revised opinion that has been reviewed and approved by the reviewer through parsing the revised opinions.

[0149] In this embodiment, each revision opinion is parsed to separate the content before and after the revision. Keywords are extracted from the content before and after the revision using text mining or natural language processing technology. These keywords should match or be related to the pre-set keywords as much as possible.

[0150] S27, determining the similarity with the contract clauses in the plurality of contract texts according to the keywords corresponding to each revision opinion and the levels corresponding to the keywords.

[0151] In this embodiment, the keywords extracted from the revision opinions are matched with the keywords in the contract clauses to calculate the similarity between them. The similarity can be measured by the overlap of keywords, semantic similarity, or other custom algorithms. When calculating the similarity, the level of keywords is taken into account, and higher weights are given to keywords with higher levels. This helps ensure that more important clauses are given priority in the matching process.

[0152] Specifically, the keyword similarity matches, such as the revision opinion is "modify 12345678", and the proposed modified clause is "modify 00345678", the character overlap rate between this revision opinion and the proposed modified clause is 80%.

[0153] S28, matching the contract clauses associated with each approved revision opinion in multiple contract texts according to similarity, and displaying the revision opinions in descending order according to the levels corresponding to the keywords in the contract clauses.

[0154] Based on the similarity score, match the contract clauses associated with each approved revision opinion in multiple contract texts. The matching process should take into account the context and overall structure of the clause to ensure the accuracy of the match. Sort the matched contract clauses in order of similarity from high to low and display them to the revision reviewer. The display interface should be clear and easy for the reviewer to quickly browse and evaluate the matching results.

[0155] In addition to modifying the proposed revised clauses, this implementation can also use the revision opinions to compare all clauses of all contracts in the text library, and arrange the clause character overlap rates from high to low. Revision reviewers can check whether other contract texts need to be revised synchronously based on the character overlap rate, thereby improving the overall coordination of revision.

[0156] It should be noted that for the above screening and revision operations, this method supports batch and individual personalized selection and revision, and the clauses in the contract text library are revised on demand.

[0157] In this embodiment, through the structured contract text revision method, the proposed revised clauses can be matched with the clauses in all contract texts for keyword similarity, key words can be set, similar clauses can be screened and displayed, and clauses with high similarity in other contract texts can be visually selected according to needs for unified revision.

[0158] The embodiments of the present application also provide corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.

[0159] The device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the contract revision method described in any embodiment of the present application.

[0160] The computer storage medium stores codes, and when the codes are executed, the device executing the codes implements the contract revision method described in any embodiment of the present application.

[0161] The "first" and "second" in the names such as "first" and "second" (if any) mentioned in the embodiments of the present application are only used as name identifiers and do not represent the first or second in order.

[0162] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0163] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0164] The above description is merely an exemplary embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. A contract revision method, characterized in that: The method comprises: Obtain multiple revision opinions on multiple contract texts; In response to selecting at least one revision element of the plurality of revision opinions, the plurality of revision opinions are merged according to a preset merging rule based on the selected revision element to obtain a merged revision opinion, so as to review the merged revision opinion to obtain a revision opinion that has been reviewed and approved; the revision elements include reason category, contract text category, contract text content, content before revision, and content after revision; the preset merging rule includes one or more of content merging, revision opinion priority merging, and comprehensive merging; Parsing the approved revision opinions, and matching the contract clauses associated with the approved revision opinions in multiple contract texts; Based on each approved revision opinion, the contract clauses associated with each approved revision opinion are revised to obtain a revised contract text; The parsing of the approved revision opinions, and matching the contract clauses associated with the approved revision opinions in multiple contract texts, includes: Obtain corpus data and contract terms related to contract business scenarios; Establishing a mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms; Structuring the contract terms in the plurality of contract texts based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms; Parsing the approved revision opinions, and matching the contract terms associated with the approved revision opinions in the structured contract terms; The contract terms in the multiple contract texts are structured based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms, including: Extracting metadata from the multiple contract texts to generate a set of contract clauses; Performing syntactic dependency analysis on each contract clause in the contract clause set to obtain a syntactic dependency tree corresponding to each contract clause; Based on the syntactic dependency tree corresponding to each contract clause, the contract clauses in multiple contract texts are structured using the mapping relationship between the corpus data corresponding to the contract business scenario and the contract clauses to obtain structured contract clauses.

2. The contract revision method according to claim 1, characterized in that: When the preset merging rule is content merging, merging the multiple revision opinions according to the preset merging rule to obtain the merged revision opinions includes: Extracting entities from the revised contents of the plurality of revision opinions; Based on the merging priority corresponding to each entity in the revised content of the multiple revision opinions, the multiple revision opinions are merged to obtain merged revision opinions. The merging priority is used to determine the merging method for merging similar revision opinions. The similar revision opinions are calculated based on the similarity of the multiple revision opinions.

3. The contract revision method according to claim 1, characterized in that: The acquisition of corpus data and contract terms related to the contract business scenario includes: Build the contract text management library, procurement type management library, material type management library, contract clause management library, and attribute tag management library corresponding to the business scenarios; Based on the contract text management library, procurement type management library, material type management library, contract clause management library and attribute label management library corresponding to the business scenario, corpus data and contract clauses related to the business scenario are set respectively.

4. The contract revision method according to claim 1, characterized in that: The method further comprises: Pre-set the keywords in the contract terms and the corresponding levels of the keywords; By parsing the revision opinions, keywords are extracted from the pre-revision content and the post-revision content of each revision opinion; Determine the similarity with the contract clauses in the multiple contract texts according to the keywords corresponding to each revision opinion and the levels corresponding to the keywords; The contract clauses associated with the revision opinions are matched in multiple contract texts according to the similarity, and the revision opinions are displayed in a descending order according to the levels corresponding to the keywords in the contract clauses.

5. The contract revision method according to claim 1, characterized in that: The contract clauses associated with each approved revision opinion are revised based on each approved revision opinion to obtain a revised contract text, including: Locate the contract clauses associated with each approved revision opinion in the contract text based on each approved revision opinion; According to the revised content of the revision opinions, the contract clauses associated with each approved revision opinion are revised in batches to obtain multiple revised contract texts; A revision report is generated based on the contract text before revision and the corresponding contract text after revision; the revision report at least includes revision information and a revision difference comparison table.

6. A contract revision system, characterized in that: The system includes a contract revision opinion collection module, a revision opinion business review module and a revision opinion confirmation module; The contract revision opinion collection module is used to obtain multiple revision opinions on multiple contract texts; A revision opinion business review module, configured to, in response to selecting at least one revision element of the plurality of revision opinions, merge the plurality of revision opinions according to a preset merging rule based on the selected revision element to obtain a merged revision opinion, so as to review the merged revision opinion to obtain a revision opinion that has been reviewed and approved; the revision elements include reason category, contract text category, contract text content, content before revision, and content after revision; The preset merging rules include one or more of content merging, revision opinion priority merging and comprehensive merging; A revision opinion confirmation module is used to parse the approved revision opinions, match the contract clauses associated with the approved revision opinions in multiple contract texts; and revise the contract clauses associated with each approved revision opinion based on each approved revision opinion to obtain a revised contract text; The revision opinion confirmation module is specifically used for: Obtain corpus data and contract terms related to contract business scenarios; Establishing a mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms; Structuring the contract terms in the plurality of contract texts based on the mapping relationship between the corpus data corresponding to the contract business scenario and the contract terms to obtain structured contract terms; Parsing the approved revision opinions, and matching the contract terms associated with the approved revision opinions in the structured contract terms; The revision opinion confirmation module is specifically used for: Extracting metadata from the multiple contract texts to generate a set of contract clauses; Performing syntactic dependency analysis on each contract clause in the contract clause set to obtain a syntactic dependency tree corresponding to each contract clause; Based on the syntactic dependency tree corresponding to each contract clause, the contract clauses in multiple contract texts are structured using the mapping relationship between the corpus data corresponding to the contract business scenario and the contract clauses to obtain structured contract clauses.

7. An electronic device, comprising a memory and a processor, wherein the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes, so that the device executes the contract revision method described in any one of claims 1 to 5.

8. A computer storage medium storing codes, wherein when the codes are executed, a device executing the codes implements the contract revision method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Contract risk approval method and device, electronic equipment and storage medium

    CN117974025A

  • System and method for creating and executing data-driven legal contracts

    US20170287090A1

  • Multi-policy processing of a document

    US20210209708A1