Intelligent electronic contract auditing and risk early warning method based on AI

Abstract

The invention provides an intelligent electronic contract auditing and risk early warning method based on AI, and belongs to the field of contract auditing, and the method comprises the following steps: building a standard database for contract auditing, carrying out the pre-classification of contract types, selecting one contract from contracts of the same type as a standard template, and carrying out the risk early warning of the contracts; the standard templates are imported into a standard database to serve as templates, and each input standard template is classified and marked; a general auditing process and an independent auditing process are set for each type of contract, the general auditing process is used for comparing and verifying each preset item in the standard template with the contract signed by adopting the standard template, and target marking is carried out on distinguishing points in the general auditing process; wherein in the independent auditing process, terms added in a standard template are subjected to cross auditing, similar contracts of the same kind are selected through classification marking, and the whole contract is traversed to select similar terms for comparison auditing.

Description

Technical Field

[0001] The present invention belongs to the field of contract review, and in particular to an AI-based intelligent electronic contract review and risk warning method. Background Art

[0002] At present, with the advent of the digital age, electronic contracts have become more and more common in commercial transactions. Their convenience and efficiency have led more and more companies to choose to use electronic contracts for business transactions. However, the review and management of electronic contracts also face many challenges, especially in terms of contract compliance, risk identification and content review. Traditional manual review methods have exposed a series of defects.

[0003] Traditional contract audits often rely on manual inspections, which is not only time-consuming and labor-intensive, but also easily subject to the subjective judgment and experience of auditors, resulting in inconsistent and unreliable audit results. Manual audits are prone to omissions when faced with a large number of contracts, especially those with complex clauses and many details, which in turn leads to potential legal risks and economic losses.

[0004] Existing contract review systems usually lack intelligent and automated functions and cannot process contract data efficiently. Manually extracting key information and clauses from a large number of contract documents is cumbersome and error-prone, and cannot quickly respond to companies' needs for contract review. This inefficiency prolongs the contract execution cycle and affects the company's business progress.

[0005] Traditional audit methods often lack the ability to identify risks in contracts. Potential legal risks, breach of contract clauses or unfair clauses in many contracts are difficult to be discovered in a timely manner through manual review, which brings serious legal consequences to enterprises. Contract reviewers often lack sufficient professional knowledge and experience, resulting in inaccurate interpretation of complex legal clauses and increasing uncertainty in contract execution. In addition, current technology also has obvious shortcomings in contract version management and tracking. Many companies find it difficult to effectively manage different versions of contracts during contract modification and changes, resulting in information confusion and increasing the likelihood of disputes.

[0006] The existing contract review system often cannot be effectively integrated with other management systems of the enterprise, resulting in the generation of data islands, which affects the circulation and sharing of information. With the continuous improvement of enterprises' requirements for the efficiency and accuracy of contract review, the intelligent electronic contract review and risk warning method based on artificial intelligence has emerged. By introducing advanced technologies such as machine learning, natural language processing and data mining, this method can realize the automated review and risk identification of electronic contracts, greatly improving the efficiency and accuracy of contract review. The intelligent contract review system can quickly parse the contract text, extract key information, and conduct risk assessment through algorithms to provide enterprises with real-time risk warning and decision support. This new review method can not only reduce the workload of manual review and improve the accuracy of contract review, but also effectively reduce the legal risks of enterprises in the process of contract execution and promote the process of digital transformation of enterprises. In summary, the emergence of intelligent electronic contract review and risk warning methods based on AI provides an effective solution to solve the various defects in traditional contract review, helping enterprises to improve the efficiency and security of contract management in the fierce market competition. Now there is an urgent need for intelligent electronic contract review and risk warning methods based on AI. Summary of the invention

[0007] The present invention proposes an AI-based intelligent electronic contract review and risk warning method, which solves the problems of low efficiency and insufficient accuracy of contract review in the prior art. By establishing a standard database and a classified review process, it realizes automated comparison, semantic analysis and highlighting, and timely identifies potential risks and ambiguities in contracts, thereby improving the accuracy and security of contract review and optimizing the process and resource allocation of manual review.

[0008] The technical solution of the present invention is implemented as follows: an AI-based intelligent electronic contract review and risk warning method, the method comprising the following steps:

[0009] Establish a standard database for contract review, pre-classify contract types, select a contract from the same type of contracts as a standard template, import the standard template into the standard database as a template, and classify and mark each entered standard template;

[0010] For each type of contract, a general review process and an independent review process are set up. The general review process compares and verifies each clause preset in the standard template with the contract signed using the standard template, and marks the differences between them; the independent review process cross-reviews the clauses added to the standard template, selects similar contracts of the same type through classification marks, and traverses the entire contract to select similar clauses for comparative review;

[0011] Perform semantic analysis on the target marked text, and associate the semantics of the target marked text with the text before and after it. Perform semantic analysis on the text one by one, and set a warning threshold for the ambiguity between the target marked text and the text before and after it. When the number and degree of ambiguity exceed the set warning threshold, highlight the target marked text.

[0012] The highlighted text is manually reviewed again, and the number of manual reviewers and the process are matched according to the contract type and contract amount. For the clauses added in the independent review process, after selecting similar clauses in similar contracts, the similar clauses are extracted, and while undergoing manual review, the similar clauses are compared with the added clauses for manual verification, and the clause analysis conclusions of the extracted approximate clauses are called at the same time, and the clause analysis conclusions are synchronized to the user end for manual verification.

[0013] Compared with existing technologies, the AI-based intelligent electronic contract review and risk warning method shows significant differences and innovations. Existing technologies often rely on manual review in the contract review process, and the review standards and processes are relatively fixed, lacking flexibility and intelligence. In contrast, this method fundamentally improves the systematization and standardization of contract review by establishing a standard database for contract review, pre-classifying contract types, selecting standard templates and classifying and marking them. Through standardized contract templates, consistency in the review process can be ensured, reducing subjectivity and inconsistency caused by manual review.

[0014] The contract review process in the prior art is usually linear, lacking in-depth analysis and comparison of contract terms, while this method sets up a general review process and an independent review process. The general review process compares and verifies each clause in the standard template, and can accurately identify the differences between the contract and the standard template and mark the target. This process significantly improves the efficiency and accuracy of the review. At the same time, the independent review process cross-reviews the clauses added to the standard template and uses similar contracts for traversal comparison, which can deeply explore potential risks and problems, making up for the lack of comprehensive review of new clauses in the prior art.

[0015] When performing semantic analysis on the target marked text, this method introduces semantic analysis technology to perform correlation analysis on the semantics before and after the target marked text. This deep semantic analysis can effectively identify ambiguities in contract terms. Existing technologies usually only rely on surface text comparisons and lack understanding and judgment of the inherent meaning of the terms. By setting warning thresholds, the number and degree of ambiguity are monitored in real time to ensure that potential risks can be highlighted and processed in a timely manner when they are discovered, which greatly enhances the intelligence of contract review and risk warning capabilities.

[0016] This method conducts manual repeated review after highlighting the text, matches the number of reviewers and the process according to the contract type and contract amount, and realizes the effective combination of intelligent review and manual review. In the existing technology, manual review often cannot flexibly adjust the review resources according to the specific circumstances of the contract. This method can improve the efficiency and accuracy of the review through intelligent matching, ensuring that each contract can get the attention it deserves. At the same time, in the independent review process, the comparison of the added terms and similar terms, combined with the synchronization of the terms analysis conclusions, will effectively reduce the workload of manual review and improve the pertinence and efficiency of the review.

[0017] The AI-based intelligent electronic contract review and risk warning method shows significant differences from existing technologies by establishing a standard database, intelligent review process, deep semantic analysis, and intelligent matching of manual review. This method not only improves the efficiency and accuracy of contract review, but also effectively reduces potential legal risks, promotes the intelligent process of contract management, and provides stronger support for corporate compliance management.

[0018] As a preferred implementation, when establishing a standard contract database, targeted data information is captured from open source data websites on the Internet, the captured open source contract data is organized into a large database in the standard database, and keywords are extracted from the captured contracts and classified.

[0019] As a preferred implementation, when selecting a standard template, all contracts to be classified are first sorted out and analyzed, classified accordingly by keywords, and arranged and weighted according to preset priorities. A threshold for the frequency of occurrence of clauses is set, and the clauses involved in the contracts of each category in the standard database are counted. The clauses of the contracts in the category that exceed the set frequency threshold are recorded, and the contracts with all recorded clauses are selected, and the standard template is selected according to the weight assigned by priority.

[0020] As a preferred implementation, when performing semantic analysis on the target marked text, a feature classification set is established for the marked text data, and a contract feature framework is built based on the feature classification. The classification sets are placed accordingly according to the framework, and the text semantics are analyzed through the text data feature classification set.

[0021] As a preferred implementation, after each audit is completed, the independent audit process establishes a non-standard terms database, and separately stores the audit data of the regular terms of the classified contracts. During each comparison, a traversal comparison is performed in the non-standard terms database. If the overlapping text data is greater than the set value, the clause is audited through the general audit process. If the overlapping text data is lower than the set value, the independent audit process is used for audit.

[0022] After adopting the above technical solution, the beneficial effect of the present invention is that by establishing a standard database and pre-classifying the contract types, the method can quickly identify and process different types of contracts. Selecting a standard template as a benchmark in similar contracts allows subsequent audit work to be carried out around this standard, thereby reducing the complexity of manual audits and improving work efficiency. In this way, auditors can focus more on the key terms and potential risks of the contract, rather than blindly searching in a large number of contracts. The combination of the general audit process and the independent audit process makes the audit process more systematic and comprehensive. The general audit process can compare and verify each preset clause in the standard template, mark the differences in time, and provide auditors with clear focus points. The independent audit process conducts in-depth analysis of the clauses added to the standard template, and cross-audits by selecting similar contracts of the same type to ensure that all clauses are strictly compared and verified. This multi-level audit mechanism effectively reduces the risk of missing key terms and ensures the compliance and effectiveness of the contract.

[0023] The introduction of semantic analysis means that the review of target marked text is not limited to the comparison of surface text, but goes deep into the connotation and context of the text. This process can identify potential ambiguities and unclear expressions, set warning thresholds to determine the number and degree of ambiguity, and thus achieve in-depth understanding and analysis of the contract content. When the number and degree of ambiguity exceed the set threshold, the relevant text is highlighted to remind the auditor to pay special attention to these potential problems. This intelligent early warning mechanism greatly improves the accuracy of contract review, ensuring that potential risks are discovered and handled in a timely manner before they are resolved.

[0024] The manual repetitive review steps combine the advantages of automation and artificial intelligence, further improving the reliability of the review. By matching the appropriate reviewers and review processes according to the contract type and amount, the professionalism and pertinence of the review work are ensured. In the independent review process, similar clauses are extracted for comparison, which not only improves the efficiency of the review, but also provides sufficient reference for reviewers, helping them make more accurate judgments.

[0025] Synchronizing the clause analysis conclusions to the user end for manual verification ensures that auditors can obtain analysis results and reference information in real time, further improving the convenience and effectiveness of the audit. This AI-based intelligent audit and risk warning method not only optimizes the contract audit process and improves work efficiency, but also provides enterprises with higher security in contract management, reduces legal risks, and provides strong support for corporate compliance operations and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0027] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example:

[0030] like Figure 1 As shown, an AI-based intelligent electronic contract review and risk warning method comprises the following steps:

[0031] Establish a standard database for contract review, pre-classify contract types, select a contract from the same type of contracts as a standard template, import the standard template into the standard database as a template, and classify and mark each entered standard template;

[0032] For each type of contract, a general review process and an independent review process are set up. The general review process compares and verifies each clause preset in the standard template with the contract signed using the standard template, and marks the differences between them; the independent review process cross-reviews the clauses added to the standard template, selects similar contracts of the same type through classification marks, and traverses the entire contract to select similar clauses for comparative review;

[0033] Perform semantic analysis on the target marked text, and associate the semantics of the target marked text with the text before and after it. Perform semantic analysis on the text one by one, and set a warning threshold for the ambiguity between the target marked text and the text before and after it. When the number and degree of ambiguity exceed the set warning threshold, highlight the target marked text.

[0034] The highlighted text is manually reviewed again, and the number of manual reviewers and the process are matched according to the contract type and contract amount. For the clauses added in the independent review process, after selecting similar clauses in similar contracts, the similar clauses are extracted, and while undergoing manual review, the similar clauses are compared with the added clauses for manual verification, and the clause analysis conclusions of the extracted approximate clauses are called at the same time, and the clause analysis conclusions are synchronized to the user end for manual verification.

[0035] The working principle and operation steps of the AI-based intelligent electronic contract review and risk warning method can be described in detail as follows:

[0036] First, the method establishes a standard database, the core of which is to pre-classify contract types. The main purpose of this process is to create a standardized audit framework so that different types of contracts can be effectively managed and audited. Among the contracts of the same type, one contract is selected as a standard template and imported into the standard database, which can ensure consistency and reliability in the subsequent audit process. Each entered standard template will also be classified and marked for subsequent retrieval and use.

[0037] Next, the system sets up a general review process and an independent review process for each type of contract. The general review process is designed to compare and verify each clause preset in the standard template with the actual contract. In this way, the system can identify the differences between the contract and the standard template and mark these differences. This process not only improves the efficiency of the review, but also ensures the compliance of the contract.

[0038] At the same time, the independent review process focuses on cross-reviewing the clauses added to the standard template. Through classification and tagging, the system can select similar contracts of the same type, traverse them and select similar clauses for comparative review. The benefit of this method is that it can identify potential risk clauses and ensure the integrity and consistency of the contract.

[0039] The system will perform semantic analysis on the target marked text, and analyze the correlation between the target marked text and its surrounding semantics. The key to this process is to analyze the text one by one, especially for the ambiguous content. The system sets a warning threshold. When the number and degree of ambiguity found exceeds the threshold, the system will highlight these words to remind the auditor to pay attention to potential risks.

[0040] For highlighted text, the system will conduct manual re-review. The key to this process is to match the appropriate number of manual reviewers and processes according to the contract type and amount. This personalized review strategy ensures that the review of important contracts can receive sufficient attention and resource support.

[0041] In the independent review process, after selecting the similar terms for the added terms, the system will extract the terms. While the manual review is repeated, the similar terms are compared and verified with the added terms, and the analysis conclusions of the terms in the extracted similar terms are called. This process not only improves the accuracy of the review, but also provides data support for manual review, allowing auditors to make more informed decisions based on historical data.

[0042] The reasons for setting up and operating this method are mainly reflected in the following aspects: 1. Systematic management: By establishing a standard database and classification management, the standardization and consistency of contract review are ensured; 2. Efficiency improvement: The separation of general review process and independent review process improves review efficiency and reduces the need for manual intervention; 3. Risk identification: Through semantic analysis and ambiguity detection, the system can actively identify potential risks and enhance the compliance of the contract; 4. Flexible response: The independent review process allows flexible risk analysis and evaluation of new clauses; 5. Data-driven: Using historical data and clause analysis conclusions to support manual review, making the review decision more scientific; 6. Multi-level review: Through a multi-level review mechanism, ensure that important information can receive full attention and improve the comprehensiveness of the review. This AI-based intelligent electronic contract review and risk warning method has built an efficient and reliable contract review system through systematic data management, intelligent risk identification and flexible review process, which has important practical value and market potential. It can effectively improve the efficiency and accuracy of contract review and reduce legal risks. It is particularly suitable for application in high-risk fields such as corporate compliance and legal review.

[0043] When establishing a standard contract database, targeted data information is captured from open source data websites on the Internet, and the captured open source contract data is organized into a large database in the standard database, and keywords are extracted from the captured contracts and classified.

[0044] In the process of establishing a contract standard database, the practice of targeted data information crawling from open source data websites on the Internet has shown significant innovation and efficiency compared with existing technologies. The establishment of traditional contract standard databases often relies on manual collection or sorting of existing contracts, which is time-consuming and labor-intensive and prone to incomplete or inconsistent data. This method is not only inefficient, but also often fails to cover a wide range of contract types and clauses, resulting in insufficient richness and representativeness of the database.

[0045] By crawling open source data websites in a targeted manner, we can effectively utilize Internet resources and quickly collect a large amount of contract data. This process ensures the breadth and diversity of data sources, allowing the contract standard database to cover a wider range of contract types and practical application scenarios. This automated data crawling method significantly improves the efficiency of data collection, reduces the need for manual intervention, and thus reduces the risk of human error.

[0046] After being captured, the open source contract data is organized into a large database in a standard database, which further improves the ability to organize and manage data. By establishing a centralized database, all data can be stored systematically for subsequent retrieval, analysis and use. This centralized management method is in sharp contrast to the decentralized storage in the existing technology, which helps to improve the efficiency and consistency of data management.

[0047] The practice of extracting keywords from the captured contracts for classification is an important innovation. Traditional methods often lack in-depth analysis of the contract content, making it difficult to extract and utilize key information. However, through keyword extraction and classification, important terms and information in the contract can be systematically organized, allowing the database to quickly respond to specific queries and needs in subsequent use. This keyword-based classification method not only improves the efficiency of information retrieval, but also lays a solid foundation for subsequent contract review and risk assessment.

[0048] The reasons for setting up and operating this method are mainly reflected in the following aspects: 1. Efficient use of resources: By crawling open source data through the Internet, the efficiency of data acquisition is greatly improved, and existing resources are fully utilized; 2. Data richness: Automatically crawling and organizing diverse contract data enhances the representativeness and applicability of the standard database; 3. Systematic management: Centralized storage and management of contract data facilitates subsequent retrieval and application, and improves data consistency; 4. Information extraction capabilities: Through keyword extraction and classification, it ensures that key information can be effectively identified and utilized, and improves the intelligence level of the database.

[0049] When selecting a standard template, first sort out and analyze all contracts to be classified, classify them accordingly using keywords, and arrange and weight them according to preset priorities. Set a threshold for the frequency of occurrence of clauses, count the clauses involved in each category of contracts in the standard database, record the clauses in the contracts in the category that exceed the set frequency threshold, select the contracts with all recorded clauses, and select the standard template based on the weight assigned to the priority.

[0050] Compared with the existing technology, the design of this standard template selection method has obvious innovations, especially in the systematization and intelligence of contract classification, clause analysis and template selection. Traditional contract management usually relies on manual review and classification, lacks a systematic processing flow, and easily leads to information omissions and inefficiency. This design uses keywords for corresponding classification, first combing and analyzing all contracts to be classified. This process can quickly identify and classify the main contents of the contract, thereby improving the accuracy and efficiency of classification.

[0051] On this basis, the system arranges and assigns weights according to preset priorities, which is not available in traditional methods. By setting priorities, the system can give priority to the most important or relevant clauses when processing a large number of contracts. This intelligent priority setting ensures that key clauses are not overlooked during the contract management process, enhancing the comprehensiveness and reliability of contract review. At the same time, the setting of the clause frequency threshold is another innovation of this method. By counting the frequency of occurrence of the clauses involved in each classified contract, the system can identify which clauses are common and important in the contract. This process not only optimizes the selection criteria for the clauses, but also improves the pertinence of subsequent template selection.

[0052] During the statistical process, the system will record the clauses that exceed the set frequency threshold. This data-driven approach makes the basis for selecting standard templates more scientific and objective. Traditional methods often rely on manual experience, resulting in improper selection. Through data statistics and records, the system can ensure that the selected template is more in line with the requirements of the actual contract. In addition, the system will filter out contracts containing all recorded clauses from the contract terms. This process ensures the applicability of the template and provides support for subsequent compliance checks and risk management.

[0053] When performing semantic analysis on the target marked text, a feature classification set is established for the marked text data, and a contract feature framework is built based on the feature classification. The classification sets are placed accordingly according to the framework, and the text semantics are analyzed through the text data feature classification set.

[0054] Traditional text analysis usually relies on simple keyword matching or rule-based parsing methods, which often fail to deeply understand the semantics and contextual relationships of the text, and are therefore prone to misjudgment or omissions when processing complex contract texts. This method, by establishing a feature classification set, first systematically processes the marked text data. This process can more comprehensively capture the key features and information in the text, thus laying the foundation for subsequent analysis.

[0055] In the process of building the contract feature framework, the system matches the feature classification set with the preset framework. This structured design makes text analysis no longer isolated, but integrated into a larger semantic understanding system. Compared with existing technologies, traditional methods often lack such a framework structure, resulting in one-sided and incomplete analysis results. Through this framework construction, the system can effectively organize and integrate text features of different categories, making the results of text analysis more systematic and operational.

[0056] Analyzing the semantics of text through a collection of text data feature classifications means that the system does not only focus on a single word or sentence, but understands the meaning and logical relationship of the text as a whole. This method can better capture and parse the complex terms, conditions and agreements in the contract, improving the accuracy and depth of the analysis. Traditional text analysis usually relies on a single analysis angle and lacks multi-dimensional semantic understanding, which can easily lead to the omission or misinterpretation of important information.

[0057] After each audit is completed, the independent audit process will establish a non-standard terms database, and separately store the audit data of the regular terms of the classified contracts. During each comparison, a traversal comparison will be performed in the non-standard terms database. If the overlapping text data is greater than the set value, the clause will be audited through the general audit process. If the overlapping text data is lower than the set value, the independent audit process will be used for audit.

[0058] Compared with existing technologies, the design of this independent review process is significantly innovative, especially in terms of handling non-standard terms and the flexibility of the review process. Traditional review processes usually rely on a fixed standard terms library, and all contract terms are handled in the same review process, lacking effective management of non-standard terms. This approach often leads to insufficiently detailed reviews and misses important contract risks or compliance issues. This design, by establishing a non-standard terms database, stores the review data of regular terms that are not in the classified contracts separately, significantly improving the ability to identify and manage non-standard terms.

[0059] After each review is completed, the system will record the non-standard terms in the database. This process not only helps to systematically store and manage non-standard terms, but also lays the foundation for subsequent comparative analysis. Compared with traditional methods, this separate storage method enables non-standard terms to be reviewed more professionally and carefully. Each time a comparison is made, the system traverses and compares the non-standard terms database. If the overlapping text data is greater than the set value, the clause will be reviewed through the general review process. This dynamic comparison mechanism ensures that common terms can be quickly and efficiently reviewed through the standard process, improving the review efficiency. On the contrary, if the overlapping text data is lower than the set value, an independent review process is used for review, ensuring that the review of less common or less commonly used terms is also taken seriously and corresponding review measures are taken.

[0060] This flexible audit mechanism not only improves the comprehensiveness and accuracy of the audit, but also effectively reduces the potential risks caused by non-standard clauses. Compared with existing technologies, traditional audit processes usually lack this dynamic adjustment capability, resulting in the audit of non-standard clauses being often ignored or improperly handled. This design, through refined audit processes and flexible database management, achieves a comprehensive grasp of contract terms and provides stronger protection for contract management.

[0061] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The AI-based intelligent electronic contract review and risk warning method is characterized by: The method comprises the following steps: Establish a standard database for contract review, pre-classify contract types, select a contract from the same type of contracts as a standard template, import the standard template into the standard database as a template, and classify and mark each entered standard template; For each type of contract, a general review process and an independent review process are set up. The general review process compares and verifies each clause preset in the standard template with the contract signed using the standard template, and marks the differences between them; the independent review process cross-reviews the clauses added to the standard template, selects similar contracts of the same type through classification marks, and traverses the entire contract to select similar clauses for comparative review; Perform semantic analysis on the target marked text, and associate the semantics of the target marked text with the text before and after it. Perform semantic analysis on the text one by one, and set a warning threshold for the ambiguity between the target marked text and the text before and after it. When the number and degree of ambiguity exceed the set warning threshold, highlight the target marked text. The highlighted text is manually reviewed again, and the number of manual reviewers and the process are matched according to the contract type and contract amount. For the clauses added in the independent review process, after selecting similar clauses in similar contracts, the similar clauses are extracted, and while undergoing manual review, the similar clauses are compared with the added clauses for manual verification, and the clause analysis conclusions of the extracted approximate clauses are called at the same time, and the clause analysis conclusions are synchronized to the user end for manual verification.

2. The AI-based intelligent electronic contract review and risk warning method according to claim 1, characterized in that: When establishing a standard contract database, targeted data information is captured from open source data websites on the Internet, and the captured open source contract data is organized into a large database in the standard database, and keywords are extracted from the captured contracts and classified.

3. The AI-based intelligent electronic contract review and risk warning method according to claim 1, characterized in that: When selecting a standard template, first sort out and analyze all contracts to be classified, classify them accordingly using keywords, and arrange and weight them according to preset priorities. Set a threshold for the frequency of occurrence of clauses, count the clauses involved in each category of contracts in the standard database, record the clauses in the contracts in the category that exceed the set frequency threshold, select the contracts with all recorded clauses, and select the standard template based on the weight assigned to the priority.

4. The AI-based intelligent electronic contract review and risk warning method according to claim 1, characterized in that: When performing semantic analysis on the target marked text, a feature classification set is established for the marked text data, and a contract feature framework is built based on the feature classification. The classification sets are placed accordingly according to the framework, and the text semantics are analyzed through the text data feature classification set.

5. The AI-based intelligent electronic contract review and risk warning method according to claim 1, characterized in that: After each audit is completed, the independent audit process will establish a non-standard terms database, and separately store the audit data of the regular terms of the classified contracts. During each comparison, a traversal comparison will be performed in the non-standard terms database. If the overlapping text data is greater than the set value, the clause will be audited through the general audit process. If the overlapping text data is lower than the set value, the independent audit process will be used for audit.

Images