Information processing method and device, electronic equipment and storage medium

Abstract

The invention provides an information processing method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring newly-added management information of an organization, and performing clause splitting on the newly-added management information to obtain clauses to be audited; searching and matching in a pre-established clause library according to the to-be-audited clauses to obtain a matched clause list; according to the first term type of the to-be-audited term, classifying candidate terms in the matched term list to obtain similar candidate terms and conflict candidate terms corresponding to the to-be-audited term; based on the similar candidate terms and the conflict candidate terms, generating correction guide information of the to-be-audited terms; and correcting the to-be-audited clauses according to the correction guide information to obtain target audited clauses, and updating the target audited clauses to a clause library. Therefore, according to the scheme, the comprehensiveness and efficiency of system review can be improved, and conflicts or contradictions between newly-added information and existing information are avoided.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to an information processing method, device, electronic device and storage medium. Background Art

[0002] When banks formulate management information, such as rules and regulations, they need to compare and review the systems issued by the regulatory authorities and the systems issued by the bank in the past to ensure that there are no conflicts between the current system and the regulatory rules, superior regulations and systems. Therefore, the compliance review of the bank's system is a very important part of the formulation of the system to ensure that the formulated system complies with the relevant requirements of laws, regulations, policies and regulations, industry standards, etc. Summary of the invention

[0003] The present disclosure provides an information processing method, device, electronic device and storage medium, which are used to solve the problem of low efficiency in manually comparing new clauses with historical clauses.

[0004] To this end, one objective of the present disclosure is to provide an information processing method.

[0005] A second objective of the present disclosure is to provide an information processing device.

[0006] A third objective of the present disclosure is to provide an electronic device.

[0007] A fourth object of the present disclosure is to provide a non-transitory computer-readable storage medium.

[0008] A fifth object of the present disclosure is to provide a computer program product.

[0009] To achieve the above-mentioned purpose, the first aspect of the present disclosure proposes an information processing method, including: obtaining new management information of an organization, and splitting the new management information into clauses to obtain clauses to be reviewed; searching and matching the clauses to be reviewed in a pre-established clause library to obtain a list of matching clauses; classifying the candidate clauses in the list of matching clauses according to the first clause type of the clauses to be reviewed to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed; generating correction guidance information for the clauses to be reviewed based on the similar candidate clauses and conflicting candidate clauses; and correcting the clauses to be reviewed according to the correction guidance information to obtain target review clauses, and updating them to the clause library.

[0010] According to one embodiment of the present disclosure, the to-be-reviewed clause is revised according to the revision guide information to obtain the target review clause, including: generating a big model prompt word corresponding to the to-be-reviewed clause based on the revision guide information; calling the big model, and inputting the big model prompt word and the to-be-reviewed clause into the big model, and the big model revising the to-be-reviewed clause according to the big model prompt word to obtain the target review clause.

[0011] According to one embodiment of the present disclosure, the generation of correction guidance information for the clause to be reviewed based on the similar candidate clauses and the conflicting candidate clauses includes: extracting keywords from the similar candidate clauses and the conflicting candidate clauses to obtain a first keyword set; extracting keywords from the clause to be reviewed to obtain a second keyword set; matching the first keyword set and the second keyword set, and generating correction guidance information for the clause to be reviewed based on the matching result.

[0012] According to one embodiment of the present disclosure, the candidate clauses in the matching clause list are classified according to the first clause type of the clause to be reviewed to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed, including: obtaining the second clause type of the candidate clauses in the matching clause list; comparing the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

[0013] According to one embodiment of the present disclosure, searching and matching the terms to be reviewed in a pre-established terms library to obtain a list of matching terms includes: acquiring first vector information of the terms to be reviewed; searching and matching the terms library according to the first vector information to obtain the list of matching terms.

[0014] According to one embodiment of the present disclosure, the method also includes: splitting the historical management documents and the regulatory system documents into clauses to obtain candidate clauses; obtaining the second vector information and clause type of the candidate clauses; and constructing the clause library based on the second vector information of the candidate clauses and the clause type of the candidate clauses.

[0015] According to one embodiment of the present disclosure, the method also includes: for any clause among the candidate clauses and the clauses to be reviewed, performing word segmentation on the any clause to obtain an initial word segmentation result; determining the word segmentation result of the any clause according to the clause type of the any clause and the initial word segmentation result; performing vector generation on the word segmentation result of the any clause to obtain vector information of the any clause; wherein, if the any clause is the clause to be reviewed, the vector information of the any clause is the first vector information; if the any clause is the candidate clause, the vector information of the any clause is the second vector information.

[0016] According to one embodiment of the present disclosure, determining the word segmentation result of any of the clauses based on the clause type of any of the clauses and the initial word segmentation result includes: in response to the clause type of any of the clauses being a prohibited clause, removing prohibited keywords from the initial word segmentation result of any of the clauses to obtain the word segmentation result of any of the clauses; in response to the clause type of any of the clauses being a compliance clause, determining the initial word segmentation result of any of the clauses to be the word segmentation result of any of the clauses.

[0017] According to one embodiment of the present disclosure, the process of determining the type of clause in any of the clauses includes: identifying prohibited keywords in any of the clauses to determine whether any of the clauses carries prohibited keywords; in response to the presence of the prohibited keywords in any of the clauses, determining that any of the clauses is a prohibitive clause; in response to the absence of the prohibited keywords in any of the clauses, determining that any of the clauses is a compliance clause.

[0018] According to one embodiment of the present disclosure, after generating the correction guidance information of the clause to be reviewed based on the similar candidate clauses and the conflicting candidate clauses, it also includes: reviewing the similar candidate clauses and the conflicting candidate clauses to obtain review results, and correcting the similar candidate clauses and the conflicting candidate clauses corresponding to the clause to be reviewed based on the review results to obtain target similar candidate clauses and target conflicting candidate clauses corresponding to the clause to be reviewed.

[0019] According to an embodiment of the present disclosure, after obtaining the target similar candidate clauses and target conflict candidate clauses corresponding to the clause to be reviewed, it also includes: generating a system revision task based on the target similar candidate clauses and target conflict candidate clauses, and sending the system revision task to the second terminal.

[0020] To achieve the above-mentioned purpose, the second aspect of the present disclosure proposes an information processing device, including: a splitting module, used to obtain the newly added management information of the organization, and to split the newly added management information into clauses to obtain clauses to be reviewed; a matching module, used to search and match the clauses to be reviewed in a pre-established clause library to obtain a list of matching clauses; a classification module, used to classify the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed, so as to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed; a generation module, used to generate correction guidance information of the clause to be reviewed based on the similar candidate clauses and conflicting candidate clauses; a correction module, used to correct the clause to be reviewed according to the correction guidance information, obtain the target review clause, and update it to the clause library.

[0021] According to one embodiment of the present disclosure, the correction module is also used to: generate a big model prompt word corresponding to the clause to be reviewed based on the correction guidance information; call the big model, and input the big model prompt word and the clause to be reviewed into the big model, and the big model corrects the clause to be reviewed according to the big model prompt word to obtain the target review clause.

[0022] According to one embodiment of the present disclosure, the generation module is also used to: extract keywords from the similar candidate clauses and the conflicting candidate clauses to obtain a first keyword set; extract keywords from the clauses to be reviewed to obtain a second keyword set; match the first keyword set and the second keyword set, and generate correction guidance information for the clauses to be reviewed based on the matching results.

[0023] According to one embodiment of the present disclosure, the classification module is further used to: obtain the second clause type of the candidate clause in the matching clause list; compare the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

[0024] According to one embodiment of the present disclosure, the matching module is further used to: obtain first vector information of the clause to be reviewed; search and match in the clause library according to the first vector information to obtain the matching clause list.

[0025] According to one embodiment of the present disclosure, the matching module is also used to: split the historical management documents and the regulatory system documents into clauses to obtain candidate clauses; obtain the second vector information and clause type of the candidate clauses; and construct the clause library based on the second vector information of the candidate clauses and the clause type of the candidate clauses.

[0026] According to one embodiment of the present disclosure, the matching module is also used to: for any clause among the candidate clauses and the clauses to be reviewed, perform word segmentation on the any clause to obtain an initial word segmentation result; determine the word segmentation result of the any clause according to the clause type of the any clause and the initial word segmentation result; perform vector generation on the word segmentation result of the any clause to obtain vector information of the any clause; wherein, if the any clause is the clause to be reviewed, the vector information of the any clause is the first vector information; if the any clause is the candidate clause, the vector information of the any clause is the second vector information.

[0027] According to one embodiment of the present disclosure, the matching module is also used to: in response to the clause type of any of the clauses being a prohibition clause, remove the prohibited keywords from the initial word segmentation result of any of the clauses to obtain the word segmentation result of any of the clauses; in response to the clause type of any of the clauses being a compliance clause, determine that the initial word segmentation result of any of the clauses is the word segmentation result of any of the clauses.

[0028] According to one embodiment of the present disclosure, the matching module is also used to: identify prohibited keywords in any of the clauses to determine whether any of the clauses carries prohibited keywords; in response to the presence of the prohibited keywords in any of the clauses, determine that any of the clauses is a prohibitive clause; in response to the absence of the prohibited keywords in any of the clauses, determine that any of the clauses is a compliance clause.

[0029] According to one embodiment of the present disclosure, the generation module is also used to: review the similar candidate clauses and conflicting candidate clauses to obtain review results, and based on the review results, revise the similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed to obtain target similar candidate clauses and target conflicting candidate clauses corresponding to the clauses to be reviewed.

[0030] According to an embodiment of the present disclosure, the revision module is further used to: generate a system revision task according to the target similar candidate clauses and the target conflict candidate clauses, and send the system revision task to the second terminal.

[0031] To achieve the above-mentioned purpose, the third aspect embodiment of the present disclosure proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to implement the information processing method as described in the first aspect embodiment of the present disclosure.

[0032] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present disclosure proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the information processing method as described in the first aspect embodiment of the present disclosure.

[0033] To achieve the above-mentioned purpose, the fifth aspect of the present disclosure proposes a computer program product, including a computer program, which is used to implement the information processing method as described in the first aspect of the present disclosure when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0035] Figure 1 An example flow chart of an information processing method provided by an embodiment of the present disclosure;

[0036] Figure 2 An example diagram of a process for matching clauses provided in an embodiment of the present disclosure;

[0037] Figure 3 An example flowchart of another information processing method provided by an embodiment of the present disclosure;

[0038] Figure 4 An example diagram of a process for revising clauses based on newly added rules and regulations information within the bank provided in an embodiment of the present disclosure;

[0039] Figure 5 An example diagram of a process for revising clauses based on new rules and regulations information provided by a regulatory agency, provided in an embodiment of the present disclosure;

[0040] Figure 6 An example flow chart of a process of building a clause library in an information processing method provided in an embodiment of the present disclosure;

[0041] Figure 7 An example diagram of a process for building a clause library provided in an embodiment of the present disclosure;

[0042] Figure 8 An example diagram of a process for reviewing newly added rules and regulations information provided in an embodiment of the present disclosure;

[0043] Fig. 9 An exemplary structural diagram of an information processing device provided in an embodiment of the present disclosure;

[0044] Fig.10 This is a diagram showing an example structure of an electronic device provided in an embodiment of the present disclosure.

[0045] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0046] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0048] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with the provisions of relevant laws and regulations.

[0049] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards.

[0050] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the relevant content of the solution.

[0051] Figure 1 This is an example flow chart of the information processing method provided by the embodiment of the present disclosure. Figure 1 As shown, the information processing method includes:

[0052] S101, obtaining the newly added management information of the organization, and splitting the newly added management information into clauses to obtain the clauses to be reviewed.

[0053] It should be noted that the execution subject of the information processing method provided in the embodiment of the present disclosure is an electronic device, and the electronic device may be a terminal device. Optionally, the terminal device may be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device may be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device may be a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc. The embodiment of the present disclosure is not specifically limited.

[0054] In some implementations, the latest published laws, regulations, standards, normative documents and other information can be used as the newly added management information of the organization. The management information can be rules and regulations information, that is, the newly added rules and regulations information can also be formed by formulating rules and regulations.

[0055] Optionally, a pre-trained big model can be used to split the newly added management information into clauses. By inputting the newly added management information into the big model, the big model splits the content of the management information according to the preset clause splitting rules to obtain the clauses to be reviewed.

[0056] In some implementations, when splitting the newly added management information, the keywords of the clauses to be reviewed can be identified, and the clause types of the clauses to be reviewed can be determined based on the keywords. For example, if the clauses to be reviewed contain prohibited keywords such as "not allowed", "prohibited", "strictly prohibited", "violations", and "not allowed", the clauses to be reviewed are determined to be prohibited clauses, otherwise they are compliance clauses.

[0057] S102, searching and matching the terms to be reviewed in a pre-established terms library to obtain a list of matching terms.

[0058] In some implementations, a search and match can be performed in the clause library based on the vector information of the clause to be reviewed to improve the efficiency of the search and match. That is, by obtaining the first vector information of the clause to be reviewed and searching and matching in the clause library based on the first vector information, a list of matching clauses is obtained.

[0059] Optionally, the first vector information of the terms to be reviewed can be obtained by vectorizing the terms to be reviewed. For example, the first vector information can be generated using a vectorization model, such as word2vec, fasttext, bert, or other vectorization models.

[0060] In some implementations, when the large model splits the newly added management information, the terms to be reviewed can be vectorized to obtain first vector information, and the first vector information and the content and type of the terms to be reviewed can be used as the splitting results, and then the first vector information of the terms to be reviewed can be determined from the splitting results.

[0061] In some implementations, a pre-established clause library contains clause vectors corresponding to clauses of all existing rules and regulations, and search and matching can be achieved in the clause library by calculating the similarity between the first vector information and the clause vectors corresponding to the candidate clauses in the clause library.

[0062] Optionally, the matching clause list may be determined based on a similarity threshold. That is, by calculating the similarity between the first vector information and the clause vector in the clause library, and determining candidate clauses with similarities greater than the similarity threshold, a matching clause list is formed. Optionally, the cosine similarity between the first vector information and the clause vector may be calculated.

[0063] S103, classifying the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed, so as to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed.

[0064] In some implementations, the clause types of the clauses include: prohibited clauses and compliance clauses. The candidate clauses may be classified according to the first clause type of the clause to be reviewed and the second clause type of the candidate clause in the matching clause list.

[0065] Optionally, the first clause type can be compared to see if it is the same as the second clause type. If they are the same, the clause to be reviewed and the candidate clause are determined to be similar clauses; if they are different, the clause to be reviewed and the candidate clause are determined to be conflicting clauses. For example, if the first clause type is a prohibition clause and the second clause type is a compliance clause, it is determined that the clause to be reviewed and the candidate clause are in conflict.

[0066] For example, if clause A of the newly added management information conflicts with candidate clause A and is similar to candidate clause B, then the similar candidate clause corresponding to clause A is determined to be candidate clause B, and the conflicting candidate clause is determined to be candidate clause A.

[0067] like Figure 2 The flowchart of the terms matching is shown in the figure. Figure 2The clauses to be reviewed in are clause 1 and clause 2. For clause 1 and clause 2, the similarity between clause 1 and clause 2 and the candidate clauses in the clause library is calculated respectively, and it is determined whether the similarity is greater than the similarity threshold. Based on the candidate clauses whose similarity is greater than the similarity threshold, a matching clause list is generated. If the similarity is less than or equal to the similarity threshold, it is determined that clause 1 and clause 2 are not related to the candidate clauses.

[0068] Furthermore, it is determined whether the first clause type of the clause to be reviewed is the same as the second clause type of the candidate clause. If they are the same, the two are determined to be similar clauses; if they are different, the two are determined to be conflicting clauses.

[0069] S104: Generate correction guidance information for the clauses to be reviewed based on similar candidate clauses and conflicting candidate clauses.

[0070] In some implementations, by determining similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed, it is possible to determine whether the newly added management information is similar to or conflicts with the existing clauses, and then generate correction guidance information for the clauses to be reviewed based on the conflict results and similar results, thereby achieving accurate evaluation of the clauses to be reviewed, increasing the compliance and rationality of the review process, and avoiding review errors caused by subjective judgment or incomplete information.

[0071] For example, if the newly added management information is divided into clause A, clause B, clause C, and clause D, the conflicting candidate clauses corresponding to clause A and clause C, and the similar candidate clauses corresponding to clause B and clause C can be determined, and then the correction guidance information can be generated according to the conflict results and similar results. For example, if clause A and clause C in the newly added management information conflict with the existing clauses, clause A and clause C are corrected.

[0072] S105, modify the to-be-audited clauses according to the correction guidance information, obtain the target audit clauses, and update them into the clause library.

[0073] In some implementations, the to-be-reviewed clauses can be revised based on the big model according to the revision guidance information, and the revised to-be-reviewed clauses can be used as target review clauses. Then, the target review clauses can be quietly updated to the clause library, so that the target review clauses can be used as candidate clauses and searched and matched with the newly added management information.

[0074] Optionally, the correction guidance information can be input into the big model to guide the big model to understand the specific content of the modification, the reason for the modification, and the suggested modification method, and to review the terms to be reviewed one by one, so as to make specific modifications to the terms to be reviewed according to the suggestions of the correction guidance information and obtain the target audit terms.

[0075] Optionally, prompt words for the large model may be generated based on the correction guidance information, so that the large model corrects the terms to be reviewed according to the prompt words to obtain target review terms.

[0076] In the management information processing method provided by the embodiment of the present disclosure, by determining the clauses to be reviewed corresponding to the newly added management information, and searching and matching in the clause library according to the clauses to be reviewed, the candidate clauses in the clause library whose similarity with the clauses to be reviewed is greater than a threshold are determined, and a list of matching clauses is generated. Based on the clause type, it is determined whether the candidate clauses in the matching clause list conflict with the clauses to be reviewed, so as to generate correction guidance information for the clauses to be reviewed. Thus, the clauses to be reviewed can be corrected according to the correction guidance information. By searching and matching the newly added management information, the comprehensiveness and efficiency of the system review can be improved, and conflicts or contradictions between the newly added management information and the existing management information can be avoided. By reviewing management information through quantization and similarity, the scientificity and rationality of the review of rules and regulations can be enhanced.

[0077] Figure 3 This is an example flow chart of the information processing method provided by the embodiment of the present disclosure. Figure 3 As shown, the information processing method includes:

[0078] S301, acquiring the newly added management information of the organization, and splitting the newly added management information into clauses to obtain the clauses to be reviewed.

[0079] S302, searching and matching the terms to be reviewed in a pre-established terms library to obtain a list of matching terms.

[0080] The relevant contents of step S301 to step S302 can be found in the above embodiment and will not be repeated here.

[0081] S303: Obtain the second clause type of the candidate clause in the matching clause list.

[0082] S304, comparing the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

[0083] In some implementations, the difference between the to-be-reviewed clause and the candidate clause is clarified by determining the second clause type of the candidate clause in the matching clause list and comparing the first clause type with the second clause type to determine whether the first clause type is the same as the second clause type.

[0084] Optionally, if the first clause type is the same as the second clause type, the candidate clause may be determined as a similar candidate clause; if the first clause type is different from the second clause type, the candidate clause may be determined as a conflicting candidate clause.

[0085] S305: Generate correction guidance information for the clauses to be reviewed based on the similar candidate clauses and the conflicting candidate clauses.

[0086] In some implementations, in order to improve the accuracy and efficiency of clause amendments and clarify the amendment direction, amendment guidance information for the clauses to be reviewed may be generated to guide the large model to amend the clauses to be reviewed.

[0087] In some implementations, the correction guidance information may be generated based on the keywords of the similar candidate clauses and the conflicting candidate clauses. The first keyword set is obtained by extracting keywords from the similar candidate clauses and the conflicting candidate clauses.

[0088] Furthermore, keywords are extracted from the terms to be reviewed to obtain a second keyword set. The first keyword set and the second keyword set are matched, and correction guidance information for the terms to be reviewed is generated based on the matching result. In other words, if there is a matching keyword between the first keyword set and the second keyword set, the keyword can be used as correction guidance information.

[0089] In some implementations, in order to determine the accuracy of the audit process and avoid audit errors, after the correction guidance information is determined, the audited clauses can be reviewed based on the correction guidance information to further confirm similar candidate clauses and conflicting candidate clauses.

[0090] Optionally, the clauses to be reviewed may be reviewed manually. The clauses to be reviewed in the revision guidance information are determined by parsing the revision guidance information, and a manual review task is generated according to similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed, and the manual review task is sent to the first terminal.

[0091] Furthermore, after receiving the manual review task, the first terminal manually reviews the clause to be reviewed based on the manual review task. The review results are obtained by reviewing the similar candidate clauses and the conflicting candidate clauses, and the similar candidate clauses and the conflicting candidate clauses corresponding to the clause to be reviewed are modified based on the review results to obtain the target similar candidate clauses and the target conflicting candidate clauses corresponding to the clause to be reviewed.

[0092] For example, in the manual review task, the similar candidate clause corresponding to the clause to be reviewed is Clause 1, and the conflicting candidate clause is Clause 2. In the review result, the similar candidate clause corresponding to the clause to be reviewed is Clause 1, and the conflicting candidate clause is Clause 3. Then, the similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed are modified, and the target similar candidate clause can be determined to be Clause 1, and the target conflicting candidate clause can be determined to be Clause 3.

[0093] In some implementations, after the target-similar candidate clauses and the target-conflicting candidate clauses are determined, the target-similar candidate clauses and the target-conflicting candidate clauses may also be revised to ensure coordination and consistency among various systems.

[0094] Optionally, a system revision task may be generated based on the target similar candidate clauses and the target conflict candidate clauses, and the system revision task may be sent to the second terminal. Optionally, the target similar candidate clauses and the target conflict candidate clauses may be carried in the system revision task and sent to the second terminal. Optionally, revision opinions may also be generated based on the target similar candidate clauses and the target conflict candidate clauses, and a system revision task may be generated based on the revision opinions.

[0095] For example, taking the bank's rules and regulations as management information, the terms and conditions can be revised based on the new rules and regulations information added within the bank, and can also be revised based on the new rules and regulations information added by the regulatory agency. Figure 4 The flowchart shown is for revising clauses based on the newly added rules and regulations information within the bank. For the newly added rules and regulations information, the clauses to be reviewed corresponding to the newly added rules and regulations information are determined, and the clauses are matched in the clause library based on the clauses to be reviewed to determine the similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed. Then, based on the similar candidate clauses and conflicting candidate clauses, a manual review task is generated, and the manual review result is determined by manual review. Based on the manual review result, the similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed are revised to obtain the target similar candidate clauses and target conflicting candidate clauses corresponding to the clauses to be reviewed. Furthermore, based on the target similar candidate clauses and target conflicting candidate clauses, a system revision task is generated.

[0096] Figure 5 The flowchart shown is for revising clauses based on the newly added rules and regulations information from the regulatory agency. For the newly added rules and regulations information, by determining the clauses to be reviewed corresponding to the newly added rules and regulations information, and matching the clauses in the clause library based on the clauses to be reviewed, the conflict candidate clauses corresponding to the clauses to be reviewed are determined. Then, based on the conflict candidate clauses, a manual review task is generated, and the manual review result is determined by manual review. Based on the manual review result, the conflict candidate clauses corresponding to the clauses to be reviewed are revised to obtain the target conflict candidate clauses corresponding to the clauses to be reviewed. Furthermore, based on the target conflict candidate clauses, a system revision task is generated.

[0097] S306, modify the to-be-audited clauses according to the correction guidance information, obtain the target audit clauses, and update them into the clause library.

[0098] In some implementations, a large model can be used to revise the audited clauses based on the revision guidance information to improve the efficiency and accuracy of the revision, further ensuring the consistency and compliance of the clauses.

[0099] Optionally, a large model prompt word corresponding to the to-be-reviewed clause can be generated based on the correction guidance information. By calling the large model and inputting the large model prompt word and the to-be-reviewed clause into the large model, the large model corrects the to-be-reviewed clause according to the large model prompt word to obtain the target review clause.

[0100] In the management information processing method provided in the embodiment of the present disclosure, by determining the clauses to be reviewed corresponding to the newly added management information, and searching and matching in the clause library based on the first vector information of the clauses to be reviewed, the candidate clauses in the clause library whose similarity with the clauses to be reviewed is greater than a threshold are determined, and a list of matching clauses is generated. Based on the clause type, it is determined whether the candidate clauses in the list of matching clauses conflict with the clauses to be reviewed, so as to generate correction guidance information for the clauses to be reviewed. Thus, a large model can be used to correct the clauses to be reviewed according to the correction guidance information. By searching and matching the newly added management information, the comprehensiveness and efficiency of the system review can be improved, and conflicts or contradictions between the newly added management information and the existing management information can be avoided. By reviewing management information by vectorization and similarity, the scientificity and rationality of the review of rules and regulations can be enhanced.

[0101] Based on the above embodiments, the present disclosure can explain the process of building a clause library, such as Figure 6 As shown, the process of building a clause library includes:

[0102] S601, split the historical management documents and regulatory system documents into clauses to obtain candidate clauses.

[0103] In some implementations, a pre-trained clause splitting model can be used to split historical rules and regulations documents and regulatory system documents to obtain candidate clauses corresponding to the historical rules and regulations documents and regulatory system documents.

[0104] S602: Obtain second vector information and clause type of the candidate clause.

[0105] In some implementations, the clause type of the candidate clause can be identified based on keywords, where the clause type includes prohibited clauses and compliance clauses. That is, if the candidate clause contains prohibited keywords, the clause type of the candidate clause is determined to be a prohibited clause; if the candidate clause does not contain prohibited keywords, the clause type of the candidate clause is determined to be a compliance clause. Among them, prohibited keywords include: not allowed, prohibited, strictly prohibited, illegal, not allowed, etc.

[0106] In some implementations, the second vector information of the candidate terms can be obtained by segmenting the candidate terms and vectorizing the term segmentation. Optionally, the second vector information can be generated using a vectorization model.

[0107] It is understandable that the method for determining the second vector information and the type of the candidate clause is the same as the method for determining the first vector information and the type of the clause to be reviewed. By determining the vectorized information of the clause, the efficiency of data reading and processing can be improved, which is conducive to the rapid search and matching of the clause and saves the space for clause storage.

[0108] That is to say, for any clause among the candidate clauses and the clauses to be reviewed, any clause is segmented to obtain an initial segmentation result, and the segmentation result of any clause is determined based on the clause type and the initial segmentation result of any clause.

[0109] Optionally, the initial word segmentation result can be adjusted according to the clause type to obtain the final word segmentation result. In response to the clause type of any clause being a prohibited clause, the prohibited keywords are removed from the initial word segmentation result of any clause to obtain the word segmentation result of any clause; in response to the clause type of any clause being a compliance clause, the initial word segmentation result of any clause is determined to be the word segmentation result of any clause. Removing prohibited keywords helps avoid potential legal risks and compliance issues, and reduces ambiguity and ambiguity in the clauses.

[0110] Furthermore, vectors can be generated for the word segmentation results of any clause to obtain vector information of any clause. Wherein, if any clause is a clause to be reviewed, the vector information of any clause is the first vector information; if any clause is a candidate clause, the vector information of any clause is the second vector information.

[0111] In some implementations, for any of the candidate clauses and the clauses to be reviewed, by identifying the prohibited keywords in any clause, it is determined whether any clause contains prohibited keywords. In response to the presence of prohibited keywords in any clause, any clause is determined to be a prohibited clause; in response to the absence of prohibited keywords in any clause, any clause is determined to be a compliance clause. Determining the clause type based on prohibited keywords helps to quickly identify and classify clauses, improves the efficiency of determining the clause type, and facilitates clarifying the purpose of the clause.

[0112] S603: construct a clause library according to the second vector information of the candidate clause and the clause type of the candidate clause.

[0113] In some implementations, the second vector information of the candidate clause and the clause type of the candidate clause may be associated to determine the clause type corresponding to the second vector information, and the second vector information and the clause type may be stored in an initialized database to obtain a clause library. For example, Elasticsearch may be used as the initialized database.

[0114] Figure 7 The figure shows the process flow of building a clause library. For historical management documents and regulatory system documents, a large model is used to split the clauses, obtain multiple candidate clauses, and identify the candidate clauses according to the prohibited keywords to determine whether the candidate clauses are prohibited clauses or compliance clauses.

[0115] For candidate clauses of compliance clauses, the candidate clauses are segmented to obtain initial segmentation results, and the initial segmentation results are vectorized to obtain the second vector information corresponding to the candidate clauses. For candidate clauses of prohibition clauses, the candidate clauses are segmented to obtain initial segmentation results, and the prohibition keywords are removed from the initial segmentation results to obtain the final segmentation results, and then the segmentation results are vectorized to obtain the second vector information, and then the clause library can be constructed based on the second vector information and clause type.

[0116] In the information processing method provided by the embodiment of the present disclosure, by splitting the historical management file and the regulatory system file, candidate clauses are obtained, and the second vector information and clause type corresponding to the candidate clauses are determined, so as to construct a clause library according to the second vector information and the clause type, and generate the clause library according to the vector information of the clause, which can greatly improve the efficiency of clause retrieval and matching. When generating vector information, taking out the class keyword can improve the accuracy of clause retrieval.

[0117] Taking rules and regulations information as management information as an example, Figure 8 Shown is a flow chart for reviewing newly added rules and regulations information. Figure 8 It includes: clause library management module, clause comparison module and system compliance inspection module.

[0118] In the clause library management module, by splitting the clauses of historical rules and regulations documents and regulatory system documents, candidate clauses corresponding to the historical rules and regulations documents and regulatory system documents can be obtained. After manual review and confirmation of the candidate clauses, the second vector information and clause type of the candidate clauses are determined, and a clause library is generated based on the second vector information and clause type.

[0119] In the clause comparison module, the newly added rules and regulations information is split into clauses to obtain the clauses to be reviewed, and based on the clauses to be reviewed, a search and match is performed in the clause library to determine the matching results between the clauses to be reviewed and the candidate clauses, that is, the matching clause list. The matching clause list contains candidate clauses that match the clauses to be reviewed.

[0120] In the system compliance check module, the second clause type of the candidate clause in the matching clause list is determined, and the clauses are compared based on the first clause type of the clause to be reviewed and the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses. Further, based on the similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed, the audit result information of the newly added rules and regulations information is generated, and the similar candidate clauses and conflicting candidate clauses are manually reviewed based on the audit result information, and the system revision task is determined based on the manual review results.

[0121] Corresponding to the information processing methods provided in the above-mentioned embodiments, an embodiment of the present disclosure further provides an information processing device. Since the information processing device provided in the embodiment of the present disclosure corresponds to the information processing methods provided in the above-mentioned embodiments, the implementation methods of the above-mentioned information processing methods are also applicable to the information processing device provided in the embodiment of the present disclosure and will not be described in detail in the following embodiments.

[0122] Fig. 9 This is a diagram showing an example of the structure of an information processing device provided by an embodiment of the present disclosure. Fig. 9 As shown, the information processing device includes: a splitting module 901, a matching module 902, a classification module 903, a generation module 904 and a correction module 905.

[0123] The splitting module 901 is used to obtain the newly added management information of the organization, and split the newly added management information into clauses to obtain the clauses to be reviewed;

[0124] A matching module 902 is used to search and match the terms to be reviewed in a pre-established terms library to obtain a list of matching terms;

[0125] A classification module 903 is used to classify the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed, so as to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed;

[0126] A generating module 904, configured to generate correction guidance information of the to-be-reviewed clause based on the similar candidate clauses and the conflicting candidate clauses;

[0127] The correction module 905 is used to correct the to-be-reviewed clause according to the correction guidance information, obtain the target review clause, and update it into the clause library.

[0128] According to one embodiment of the present disclosure, the correction module 905 is also used to: generate a big model prompt word corresponding to the clause to be reviewed based on the correction guidance information; call the big model, and input the big model prompt word and the clause to be reviewed into the big model, and the big model corrects the clause to be reviewed according to the big model prompt word to obtain the target review clause.

[0129] According to one embodiment of the present disclosure, the generation module 904 is also used to: extract keywords from the similar candidate clauses and the conflicting candidate clauses to obtain a first keyword set; extract keywords from the clauses to be reviewed to obtain a second keyword set; match the first keyword set and the second keyword set, and generate correction guidance information for the clauses to be reviewed based on the matching results.

[0130] According to one embodiment of the present disclosure, the classification module 903 is further used to: obtain the second clause type of the candidate clause in the matching clause list; compare the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

[0131] According to an embodiment of the present disclosure, the matching module 902 is further used to: obtain the first vector information of the clause to be reviewed; and perform search and matching in the clause library according to the first vector information to obtain the matching clause list.

[0132] According to one embodiment of the present disclosure, the matching module 902 is also used to: split the historical management files and the regulatory system files into clauses to obtain candidate clauses; obtain the second vector information and clause type of the candidate clauses; and construct the clause library based on the second vector information of the candidate clauses and the clause type of the candidate clauses.

[0133] According to one embodiment of the present disclosure, the matching module 902 is also used to: for any clause among the candidate clauses and the clauses to be reviewed, perform word segmentation on the any clause to obtain an initial word segmentation result; determine the word segmentation result of the any clause according to the clause type of the any clause and the initial word segmentation result; perform vector generation on the word segmentation result of the any clause to obtain vector information of the any clause; wherein, if the any clause is the clause to be reviewed, the vector information of the any clause is the first vector information; if the any clause is the candidate clause, the vector information of the any clause is the second vector information.

[0134] According to one embodiment of the present disclosure, the matching module 902 is also used for: in response to the clause type of any of the clauses being a prohibition clause, removing the prohibited keywords from the initial word segmentation result of any of the clauses to obtain the word segmentation result of any of the clauses; in response to the clause type of any of the clauses being a compliance clause, determining that the initial word segmentation result of any of the clauses is the word segmentation result of any of the clauses.

[0135] According to one embodiment of the present disclosure, the matching module 902 is also used to: identify prohibited keywords in any of the clauses to determine whether any of the clauses carries prohibited keywords; in response to the presence of the prohibited keywords in any of the clauses, determine that any of the clauses is a prohibitive clause; in response to the absence of the prohibited keywords in any of the clauses, determine that any of the clauses is a compliance clause.

[0136] According to one embodiment of the present disclosure, the generation module 904 is also used to: review the similar candidate clauses and conflicting candidate clauses to obtain review results, and based on the review results, modify the similar candidate clauses and conflicting candidate clauses corresponding to the clauses to be reviewed to obtain target similar candidate clauses and target conflicting candidate clauses corresponding to the clauses to be reviewed.

[0137] According to an embodiment of the present disclosure, the correction module 905 is further used to: generate a system revision task according to the target similar candidate clauses and the target conflict candidate clauses, and send the system revision task to the second terminal.

[0138] In the management information processing device provided by the embodiment of the present disclosure, by determining the clauses to be reviewed corresponding to the newly added management information, and searching and matching in the clause library according to the clauses to be reviewed, the candidate clauses in the clause library whose similarity with the clauses to be reviewed is greater than a threshold are determined, and a list of matching clauses is generated. Based on the clause type, it is determined whether the candidate clauses in the matching clause list conflict with the clauses to be reviewed, so as to generate correction guidance information for the clauses to be reviewed. Thus, the clauses to be reviewed can be corrected according to the correction guidance information. By searching and matching the newly added management information, the comprehensiveness and efficiency of the system review can be improved, and conflicts or contradictions between the newly added management information and the existing management information can be avoided. By reviewing management information through quantization and similarity, the scientificity and rationality of the review of rules and regulations can be enhanced.

[0139] Fig.10 This is a diagram showing an example of the structure of an electronic device provided in an embodiment of the present disclosure. Fig.10 As shown, the electronic device 1000 may include: a transceiver 1001 , a processor 1002 , and a memory 1003 .

[0140] The processor 1002 executes the computer execution instructions stored in the memory, so that the processor 1002 executes the scheme in the above embodiment. The processor 1002 can be a general-purpose processor, including a central processing unit CPU, a network processor (network processor, NP), etc.; it can also be a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0141] The memory 1003 is connected to the processor 1002 via a system bus and completes communication between them. The memory 1003 is used to store computer program instructions.

[0142] The transceiver 1001 may be used to obtain tasks to be executed and configuration information of the tasks to be executed.

[0143] The system bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The transceiver is used to realize the communication between the database access device and other computers (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM) and may also include non-volatile memory.

[0144] The electronic device provided in the embodiments of the present disclosure may be the terminal device of the above embodiments.

[0145] The disclosed embodiment also provides a chip for executing instructions, which is used to execute the technical solution of the information processing method in the above embodiment.

[0146] The embodiment of the present disclosure further provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the computer executes the technical solution of the information processing method of the above embodiment.

[0147] The embodiments of the present disclosure also provide a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, the technical solution of the information processing method in the above embodiments can be implemented.

[0148] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0149] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An information processing method, characterized in that: include: Acquire the newly added management information of the organization, and split the newly added management information into clauses to obtain the clauses to be reviewed; Search and match the terms to be reviewed in a pre-established terms library to obtain a list of matching terms; Classifying the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed, so as to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed; Based on the similar candidate clauses and the conflicting candidate clauses, generating correction guidance information for the clause to be reviewed; The clause to be reviewed is revised according to the revision guidance information to obtain the target review clause, and is updated into the clause library.

2. The method according to claim 1, characterized in that The step of modifying the to-be-reviewed clause according to the modification guidance information to obtain the target review clause includes: Generate a large model prompt word corresponding to the clause to be reviewed based on the correction guidance information; The big model is called, and the big model prompt word and the clause to be reviewed are input into the big model, and the big model amends the clause to be reviewed according to the big model prompt word to obtain the target review clause.

3. The method according to claim 2, characterized in that The step of generating correction guidance information of the clause to be reviewed based on the similar candidate clauses and the conflicting candidate clauses includes: Extract keywords from the similar candidate clauses and the conflicting candidate clauses to obtain a first keyword set; Extract keywords from the terms to be reviewed to obtain a second keyword set; The first keyword set and the second keyword set are matched, and correction guidance information of the to-be-reviewed clause is generated based on the matching result.

4. The method according to claim 1, characterized in that: The classifying the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed includes: Obtaining a second clause type of a candidate clause in the matching clause list; Compare the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

5. The method according to claim 1, characterized in that The searching and matching of the terms to be reviewed in a pre-established terms library to obtain a list of matching terms includes: Obtaining first vector information of the clause to be reviewed; A search and match is performed in the clause library according to the first vector information to obtain the matching clause list.

6. The method according to claim 1, characterized in that The method further comprises: Split the clauses of historical management documents and regulatory system documents to obtain candidate clauses; Obtaining second vector information and clause type of the candidate clause; The clause library is constructed according to the second vector information of the candidate clause and the clause type of the candidate clause.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: For any clause among the candidate clauses and the clauses to be reviewed, segment the clause to obtain an initial segmentation result; Determine a word segmentation result of any of the clauses according to the clause type of any of the clauses and the initial word segmentation result; Generate a vector for the word segmentation result of any of the clauses to obtain vector information of any of the clauses; Among them, any of the clauses is the clause to be reviewed, and the vector information of any of the clauses is the first vector information; any of the clauses is the candidate clause, and the vector information of any of the clauses is the second vector information.

8. The method according to claim 7, characterized in that Determining the word segmentation result of any of the clauses according to the clause type of any of the clauses and the initial word segmentation result includes: In response to the clause type of any of the clauses being a prohibited clause, removing prohibited keywords from an initial word segmentation result of any of the clauses to obtain a word segmentation result of any of the clauses; In response to the clause type of any of the clauses being a compliance-type clause, determining an initial word segmentation result of any of the clauses as the word segmentation result of any of the clauses.

9. The method according to claim 8, characterized in that The process of determining the type of clause in any of the above clauses includes: Identify prohibited keywords in any of the clauses to determine whether any of the clauses contains prohibited keywords; In response to the presence of the prohibited keyword in any of the clauses, determining that any of the clauses is a prohibited clause; In response to the prohibition-type keyword not existing in any of the clauses, it is determined that any of the clauses is a compliance-type clause.

10. The method according to claim 1, characterized in that After generating the correction guidance information of the clause to be reviewed based on the similar candidate clauses and the conflicting candidate clauses, the method further includes: The similar candidate clauses and conflicting candidate clauses are reviewed to obtain review results, and based on the review results, the similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed are revised to obtain target similar candidate clauses and target conflicting candidate clauses corresponding to the clause to be reviewed.

11. The method according to claim 10, characterized in that After obtaining the target similar candidate clauses and target conflict candidate clauses corresponding to the clause to be reviewed, the method further includes: A system revision task is generated according to the target-similar candidate clauses and the target-conflicting candidate clauses, and the system revision task is sent to the second terminal.

12. An information processing device, characterized in that: include: A splitting module is used to obtain the newly added management information of the organization and split the newly added management information into clauses to obtain the clauses to be reviewed; A matching module, used to search and match the terms to be reviewed in a pre-established terms library to obtain a list of matching terms; A classification module, used for classifying the candidate clauses in the matching clause list according to the first clause type of the clause to be reviewed, so as to obtain similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed; A generating module, used for generating correction guidance information of the clause to be reviewed based on the similar candidate clauses and the conflicting candidate clauses; A correction module is used to correct the clause to be reviewed according to the correction guidance information, obtain the target review clause, and update it into the clause library.

13. The device according to claim 12, characterized in that The correction module is further used for: Generate a large model prompt word corresponding to the clause to be reviewed based on the correction guidance information; The big model is called, and the big model prompt word and the clause to be reviewed are input into the big model, and the big model amends the clause to be reviewed according to the big model prompt word to obtain the target review clause.

14. The device according to claim 13, characterized in that The generating module is further used for: Extract keywords from the similar candidate clauses and the conflicting candidate clauses to obtain a first keyword set; Extract keywords from the terms to be reviewed to obtain a second keyword set; The first keyword set and the second keyword set are matched, and correction guidance information of the to-be-reviewed clause is generated based on the matching result.

15. The device according to claim 12, characterized in that The classification module is further used for: Obtaining a second clause type of a candidate clause in the matching clause list; Compare the first clause type with the second clause type of the candidate clause to obtain similar candidate clauses and conflicting candidate clauses.

16. The device according to claim 12, characterized in that The matching module is further used for: Obtaining first vector information of the clause to be reviewed; A search and match is performed in the clause library according to the first vector information to obtain the matching clause list.

17. The device according to claim 12, characterized in that The matching module is further used for: Split the clauses of historical management documents and regulatory system documents to obtain candidate clauses; Obtaining second vector information and clause type of the candidate clause; The clause library is constructed according to the second vector information of the candidate clause and the clause type of the candidate clause.

18. The device according to any one of claims 12 to 17, characterized in that The matching module is further used for: For any clause among the candidate clauses and the clauses to be reviewed, segment the clause to obtain an initial segmentation result; Determine a word segmentation result of any of the clauses according to the clause type of any of the clauses and the initial word segmentation result; Generate a vector for the word segmentation result of any of the clauses to obtain vector information of any of the clauses; Among them, any of the clauses is the clause to be reviewed, and the vector information of any of the clauses is the first vector information; any of the clauses is the candidate clause, and the vector information of any of the clauses is the second vector information.

19. The device according to claim 18, characterized in that The matching module is further used for: In response to the clause type of any of the clauses being a prohibited clause, removing prohibited keywords from an initial word segmentation result of any of the clauses to obtain a word segmentation result of any of the clauses; In response to the clause type of any of the clauses being a compliance-type clause, determining an initial word segmentation result of any of the clauses as the word segmentation result of any of the clauses.

20. The device according to claim 19, characterized in that The matching module is further used for: Identify prohibited keywords in any of the clauses to determine whether any of the clauses contains prohibited keywords; In response to the presence of the prohibited keyword in any of the clauses, determining that any of the clauses is a prohibited clause; In response to the prohibition-type keyword not existing in any of the clauses, it is determined that any of the clauses is a compliance-type clause.

21. The device according to claim 12, characterized in that The generating module is further used for: The similar candidate clauses and conflicting candidate clauses are reviewed to obtain review results, and based on the review results, the similar candidate clauses and conflicting candidate clauses corresponding to the clause to be reviewed are revised to obtain target similar candidate clauses and target conflicting candidate clauses corresponding to the clause to be reviewed.

22. The device according to claim 21, characterized in that The correction module is further used for: A system revision task is generated according to the target-similar candidate clauses and the target-conflicting candidate clauses, and the system revision task is sent to the second terminal.

23. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 11.

24. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 11 when executed by a processor.

25. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 11 when being executed by a processor.

Images