Text auditing method and device and electronic equipment

Abstract

The embodiment of the invention discloses a text auditing method and device and electronic equipment. The text auditing method comprises the following steps: under the condition that a target text to be subjected to risk auditing is received, determining a target business type to which the target text belongs from a plurality of business types; obtaining a reference line matched with the target service type and used for finding risks, wherein the reference line at least comprises an examination key point list used for representing potential risk points; determining a target review key point hit by the target text from the review key point list, and obtaining a risk scale which corresponds to the target review key point and is used for representing an acceptable risk deviation degree; and based on a risk scale corresponding to the target review key point, performing first risk review on the target text to obtain a first risk review result.

Description

Technical Field

[0001] This document belongs to the field of text review technology, and specifically relates to a method, device and electronic equipment for text review. Background Art

[0002] With the rapid development of information technology, particularly advancements in artificial intelligence and natural language processing, the review of contracts and other documents, a crucial aspect of legal and / or business activities, is undergoing profound changes. For example, traditional contract review typically requires legal professionals to review the text line by line, which is time-consuming and labor-intensive, and prone to human oversight. Furthermore, contract terms are complex and varied, and legal language is often difficult to understand, making contract review a significant challenge for non-professionals. Therefore, a better text review solution is needed to improve the accuracy of text review results. Summary of the Invention

[0003] The embodiments of this specification provide a method, device, and electronic device for text review to provide a text review solution.

[0004] In a first aspect, an embodiment of the present specification provides a method for text review, the method comprising: upon receiving a target text to be subject to risk review, determining a target business type to which the target text belongs from a plurality of business types; obtaining a baseline for discovering risks that matches the target business type, the baseline comprising at least a list of review points for characterizing potential risk points; determining the target review points hit by the target text from the list of review points, and obtaining a risk scale corresponding to the target review points for characterizing an acceptable degree of risk deviation; performing a first risk review on the target text based on the risk scale corresponding to the target review points to obtain a first risk review result.

[0005] In the second aspect, an embodiment of the present specification provides a device for text review, including: a business type module, which is used to determine, from multiple business types, the target business type to which the target text belongs when a target text to be risk reviewed is received; a baseline module, which is used to obtain a baseline for discovering risks that matches the target business type, and the baseline at least includes a list of review points for characterizing potential risk points; a risk scale module, which is used to determine the target review points hit by the target text from the list of review points, and obtain the risk scale corresponding to the target review points for characterizing the acceptable degree of risk deviation; a first review module, which is used to perform a first risk review on the target text based on the risk scale corresponding to the target review points, and obtain a first risk review result.

[0006] In a third aspect, an embodiment of the present specification provides an electronic device, which includes: a processor, and a memory arranged to store computer-executable instructions. When the executable instructions are executed, the processor is capable of: when receiving a target text to be subject to risk review, determining, from multiple business types, a target business type to which the target text belongs; obtaining a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; determining, from the list of review points, target review points hit by the target text, and obtaining a risk scale corresponding to the target review points for characterizing an acceptable degree of risk deviation; and based on the risk scale corresponding to the target review points, performing a first risk review on the target text to obtain a first risk review result.

[0007] In a fourth aspect, an embodiment of the present specification provides a storage medium for storing a computer program, where the computer program can be executed by a processor to implement the following process: when receiving a target text to be subject to risk review, determining, from multiple business types, a target business type to which the target text belongs; obtaining a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; determining, from the list of review points, target review points hit by the target text, and obtaining a risk scale corresponding to the target review points for characterizing an acceptable degree of risk deviation; and based on the risk scale corresponding to the target review points, performing a first risk review on the target text to obtain a first risk review result.

[0008] In a fifth aspect, an embodiment of the present specification provides a computer program product, including a computer program, where when the computer program is executed by a processor, the following process is implemented: when receiving a target text to be subject to risk review, determining, from multiple business types, a target business type to which the target text belongs; obtaining a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; determining, from the list of review points, target review points hit by the target text, and obtaining a risk scale corresponding to the target review points for characterizing an acceptable degree of risk deviation; and based on the risk scale corresponding to the target review points, performing a first risk review on the target text to obtain a first risk review result. Description of the Drawings

[0009] To more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments described in one or more embodiments of this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0010] Figure 1 is a schematic flowchart of a text review method provided by an embodiment of this specification; Figure 2 is a schematic overall flowchart of a text review solution provided by an embodiment of this specification; Figure 3 is a schematic diagram of a contract review process based on a baseline provided by an embodiment of this specification.

[0011] Figure 4 is a schematic structural diagram of a text review device provided by an embodiment of this specification; Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of this specification. Specific Embodiments

[0012] The following will clearly and completely describe the technical solutions in the embodiments of this specification with reference to the drawings in the embodiments of this specification. Obviously, the described embodiments are some, but not all, of the embodiments of this specification. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by this document.

[0013] The terms "first", "second", etc. in the specification and claims of this document are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this specification can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same category, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0014] The following will, with reference to the drawings, through specific embodiments and their application scenarios, detail the text review method, device, and electronic device provided by the embodiments of this specification.

[0015] Figure 1A method for text review provided by an embodiment of the present invention is shown. This method can be executed by an electronic device, which may include: a server and / or a terminal device, where the terminal device can be, for example, an in-vehicle terminal or a mobile phone terminal, etc. In other words, this method can be executed by software or hardware installed in the aforementioned electronic device. The method for text review includes the following steps: Step S102: When receiving a target text to be risk-reviewed, determine the target business type to which the target text belongs from multiple business types.

[0016] Among them, the target text is the text input by the user to be risk-reviewed, and the target text can be a contract, a specification, a report, a judicial document, etc. This specification does not specifically limit the type of the target text, which can be determined according to the actual situation. To better illustrate this specification and highlight the main idea of this specification, the specific embodiments of this specification take the text of contracts as the description object. Those skilled in the art should understand that for other texts such as reports, the solutions in this specification can also be implemented.

[0017] The business type is a further classification of the texts of each type obtained after the initial classification of the text (the classification results are the above-mentioned contracts, specifications, reports, judicial documents, etc.). For example, for the text of the contract type, the contract can be classified based on business types such as investment, procurement, and receipt, resulting in investment type contracts, procurement type contracts, receipt type contracts, etc.; for the text of the report type, the report can be classified based on business types such as sales, finance, and human resources, resulting in sales type reports, finance type reports, human resources type reports, etc. Among the texts of the same type, the content, format, etc. of the texts corresponding to different business types are often different.

[0018] In this way, when receiving the target text, the business type to which the target text belongs (i.e., the target business type) can be determined from the preset business types. In recent years, large language models have performed excellently in processing natural language tasks and can understand and generate text that complies with grammar norms and does not violate user privacy. With their powerful learning ability and rich knowledge reserve, such large language models are gradually applied to multiple fields of text processing, including the field of text review such as contracts. Through training, large language models can capture potential risks, the meaning of legal terms, and the mutual relationship between contracts in texts such as contracts, so as to provide auxiliary support for legal professionals. In addition to being directly selected by the user from the preset multiple business types, the target business type can also be automatically classified by a preset model (such as a large language model) after reading and parsing the contract content, and the target business type to which the target text belongs can be determined.

[0019] The risk review of the target text can be a compliance risk review, a legal risk review, a public opinion risk review, etc. The content of the risk review in this specification is not specifically limited and can be determined according to the actual situation.

[0020] Step S104: Obtain a baseline for risk discovery that matches the target business type.

[0021] Step S106: From the list of review key points, determine the target review key points hit by the target text, and obtain the risk scale corresponding to the target review key points for characterizing the acceptable risk deviation degree.

[0022] Among them, the baseline is a reference standard for the degree of risk determined by comprehensively considering various possible risk factors based on factors such as general knowledge and historical data of the target business type to which the target text belongs. The baseline is a "framework" structure constructed based on the rights and obligations and responsibility assignments corresponding to each link of business implementation. Based on the baseline, it can be judged whether the risk status of the target text exceeds the acceptable range. The baseline can include multiple contents such as the probability of risk occurrence, the impact of risk generation, the threshold of risk occurrence, and the time of risk occurrence.

[0023] The baseline can at least include a list of review key points for characterizing potential risk points. This list of review key points contains multiple review key points specified based on general knowledge and historical data of the business type. The review key points can be confidentiality responsibilities, liability for breach of contract, etc. After obtaining the baseline corresponding to the target business type, the target review key points hit by the target text can be determined from the list of review key points included in the baseline. In one example, a large language model can be used to determine the target review key points. Specifically, the target text and the list of review key points corresponding to the target business type to which the target text belongs can be input into the large language model together, and the large language model determines the target review key points hit by the target text to improve the accuracy of determining the target review key points corresponding to the target text.

[0024] For each review key point, there can be at least one risk scale for characterizing the acceptable risk deviation degree. This risk scale can be a specific threshold range or multiple acceptable means. For example, for the review key point of liquidated damages in a contract text, the risk scale can be 5±10% ten thousand yuan; for the review key point of notice method in a contract text, the risk scale can be email, written mailing, etc. There are many reference factors affecting the risk scale, and the risk scale can be different at different times, for different objects, and under different policy conditions. After determining the target review key points, the corresponding risk scale can be determined according to the target review key points.

[0025] Step S108: Based on the risk scale corresponding to the target review key points, conduct a first risk review on the target text to obtain a first risk review result.

[0026] Among them, the first risk review is a risk review of the target text based on the risk scale corresponding to the target review points, and the risk scale is the acceptable modification scale for the target review points.

[0027] For the target review point of liquidated damages / time, the corresponding risk scale can be the threshold range of liquidated damages / time. The first risk review can be to determine whether the actual liquidated damages amount / time in the target text exceeds the preset threshold range. For example, for the content in the target text "Any party sends a notice related to this Agreement to the other party according to the communication address listed in this Agreement. After 10 working days from the posting date, it shall be deemed that the notice has been served on the other party.", the baseline it hits can be "The change of contact information needs to be notified in advance", and the corresponding risk scale can be "Acceptable, but the notice period can be negotiated to be shortened, such as 5 working days".

[0028] For the review point of notice method, the corresponding risk scale can be multiple acceptable notice methods. The first risk review can be to determine whether the actual notice method in the target text is the preset notice method. For example, for the content in the target text "All notices related to this Agreement can be sent in the form of written mail or email.", the baseline it hits can be "Both email and mail are valid notice methods", and the corresponding risk scale can be "Acceptable, but it is necessary to ensure that the email addresses and addresses of both parties are accurate".

[0029] For the core content in the contract text, the corresponding risk scale can be 0, that is, no modification is accepted based on the preset content of the target review points. The first risk review can be to determine whether the corresponding content of the target review points in the target text is consistent with the preset content. For example, for the content in the target text "Both parties agree that Party A does not need to truthfully and fully disclose to Party B the business compliance situation and risks related to the information content of Party A's proposed access to Party B's page.", the baseline it hits can be "Party A should disclose the business compliance situation and risks, and Party B has the right to adjust the service according to the situation", and the corresponding risk scale can be "No modification is allowed. The core security responsibilities are clear, and modification may lead to compliance risks".

[0030] In the case where the target text exceeds this risk scale, it can be determined that the result of the first risk review is that there is a risk; in the case where the target text does not exceed this risk scale, it can be determined that the result of the first risk review is that there is no risk.

[0031] In the embodiments of the present specification, first, determine the target business type to which the target text belongs, and obtain a baseline that matches the target business type. Then, determine the hit target review points from the list of review points included in the baseline. Next, perform a first risk review on the target text based on the risk scale corresponding to the target review points to obtain a first risk review result. This process defines the basic scope of text review through the baseline, sets the flexible range of risk review through the risk scale in the list of review points in the baseline. The combination of the baseline and the risk scale can not only reduce the workload of manual review but also greatly improve the accuracy of risk review of the text, making the text review method have broad application prospects and market value.

[0032] Figure 2 A schematic diagram of an application scenario of a text review method is provided, as Figure 2 shown. The user sends a service request to the server through a terminal such as a mobile phone. The service request includes: performing a text review task for the target text input by the user. After receiving the service request, the server uses the text review method in this specification to obtain a risk review result corresponding to the target text and sends the risk review result to the terminal for the user's reference.

[0033] In one implementation, when the target text is a contract text, step S106 can be executed as the following steps A1 - A2: Step A1, determine the target text source to which the target text belongs from a preset text source; Step A2, based on the target text source, determine the target review points hit by the target text from the list of review points.

[0034] For text of the contract type, usually, the preset text source includes standard contracts and non - standard contracts. A standard contract is a contract generated using a template. For each type of business contract, a contract template will be preset in advance. The terms of the template are clear and the form is unified, but users are allowed to modify it. When legal affairs staff or lawyers draft a contract, they can modify and write based on this template, and the contract obtained after modification is a standard contract. A non - standard contract is a contract negotiated and formulated by the two parties entering into the contract without using a template. The content of the terms in a non - standard contract can be adjusted according to the needs and wishes of both parties, and there is no fixed format.

[0035] As described above, the text sources of standard contracts and non - standard contracts are different. The former has a unified form template, while the latter has a flexible format, which will make the determination methods of their target review points different. In one implementation, step A2 can be executed as the following steps B1 - B3: Step B1, when the target text source is a standard contract, obtain the target standard contract corresponding to the target business type; Step B2: Compare the target text with the target standard contract to identify the points for modification. Step B3: Based on the points for modification, determine the target review points hit by the target text from the list of review key points.

[0036] Among them, the target standard contract is the template contract used to construct the target text.

[0037] For a standard contract, it can be divided into a template part identical to the used template contract and a modified part different from the used template contract. The former is the part where risks have been determined not to exist, while the latter is the part where it is uncertain whether risks exist. Therefore, the determination of the target review points can be based only on the latter.

[0038] Specifically, when the source of the target text is a standard contract, the template contract (i.e., the target standard contract) used to construct the target text can be obtained based on the target business type. By comparing the target text with the target standard contract, the points for modification to the target standard contract when generating the target text can be found. Furthermore, based on the points for modification, the target review points can be determined from the list of review key points.

[0039] In the above process, when the source of the target text is a standard contract, the points for modification to the target standard contract when constructing the target text are determined, and then the target review points are determined based on the points for modification. This process determines the target review points only based on the points for modification, omitting the determination of the target review points in the non-modification part, which can improve the efficiency of the target review points hit by the target text with a standard contract source and thus improve the efficiency of text review.

[0040] For the target text whose source is a non-standard contract, the process for determining its corresponding target review points is different from the above process. In one implementation, Step A2 can be executed as the following Steps C1 - C3: Step C1: When the source of the target text is a non-standard contract, perform a splitting process on the target text to obtain multiple sub-target texts. Step C2: Determine the target review points hit by the sub-target texts from the baseline. Step C3: Perform a merging process on the target review points hit by each sub-target text included in the target text to obtain the target review points hit by the target text.

[0041] For contract texts, they contain many and dense key points, which will cause a large computational amount and low accuracy in the process of determining the target review key points in the target text. To improve the speed of the process of determining the target review key points in the target text and the accuracy of the determination results of the target review key points in the target text, the target text can be split and the target review key points can be determined for the obtained multiple sub-target texts.

[0042] In addition to dividing the target text by the number of words, since the titles in the contract text naturally divide the contract text, it is possible to determine the sub-target text based on the title. In one implementation method, step C1 can be performed as the following steps D1 - D2: Step D1, when the source of the target text is a non-standard contract, obtain the target title level used when splitting the target text. Step D2, use the target title level to split the target text to obtain multiple sub-target texts.

[0043] Among them, the target title level is the title level based on which the text is split.

[0044] When it is determined that the source of the target text is a non-standard contract, the target title level can be obtained, and then the target text can be split using the target title level to obtain multiple sub-target texts. Contracts often have a large amount of content and complex terms. The first-level titles can divide the contract content into different large sections, such as "General Provisions", "Subject Matter of the Contract", "Price and Payment", "Liability for Breach of Contract", etc., making the overall structure of the contract clear at a glance. In one example, the target title level can be set to the first level.

[0045] After obtaining the target review key points hit by the sub-target text, the target review key points hit by each sub-target text can be merged to obtain the target review key points hit by the target text. This process divides the large text (target text) into small texts (sub-target texts), making the text review process more focused and accurate.

[0046] Figure 3 Shows a schematic diagram of the contract review process based on the baseline. As Figure 3 shown, this contract review process classifies the input contract based on whether it is a standard contract and adopts different processing methods for different types of contracts. Specifically, for standard contracts, the contract clause set (i.e., the text corresponding to the target review key points hit by the target text) is obtained by obtaining the difference points, and for non-standard contracts, the contract clause set (i.e., multiple sub-target texts) is obtained by cutting and segmenting.

[0047] In the embodiments of this specification, after determining the target text source to which the target text belongs from the preset text sources, based on the target text source, the target review points are determined from the review point list. This process implements a target review point determination scheme that matches the target text with the target text source, improving the flexibility and accuracy of the process of determining the target review points.

[0048] Since the review point list is a finite set and the content of contract review is highly open-ended, in order to overcome the missed identification of some potential risks in the text review process, content other than the review point list can be added to the baseline. In one implementation, the text review method can also perform the following steps E1 - E2: Step E1, obtain business knowledge and / or legal knowledge that matches the target business type from the baseline; Step E2, use the business knowledge and / or legal knowledge to conduct a second risk review of the target text to obtain a second risk review result.

[0049] Specifically, as Figure 3 shown, the baseline can include business knowledge and / or legal knowledge. In the baselines corresponding to different business types, the business knowledge and / or legal knowledge are different. In this way, after determining the baseline that matches the target business type, the business knowledge and / or legal knowledge that matches the target business type can be obtained from the baseline. Further, the obtained business knowledge and / or legal knowledge can be used to conduct a second risk review of the target text to obtain a second risk review result.

[0050] In an example, both the second risk review process here and the aforementioned first risk review process can be carried out through a large language model. Specifically, after determining the target review points, business knowledge, and / or legal knowledge hit by the target text, the large language model can be used to conduct a risk review of the hit content to obtain the first risk review result corresponding to the review point list and the second risk review result corresponding to the business knowledge and / or legal knowledge.

[0051] It should be understood that, as Figure 3 shown, regardless of whether the target text hits the review points in the review point list, a second risk review can be conducted on the target text. The difference is that in the case of a hit, the target text needs to undergo a first risk review and a second risk review, and in the case of no hit, only a second risk review is required.

[0052] The above - mentioned second risk review process can identify potential business / legal risks through business knowledge and / or legal knowledge, making the text review process more in line with the actual contract review business, achieving a fallback for the first risk identification result, and improving the accuracy of the text review result.

[0053] Further, modification suggestions can also be specified based on the risk review results. In one implementation, the text review method can also perform the following steps F1 - F2: Step F1, determine the initial modification suggestion based on the deviation degree between the text corresponding to the target review point and the risk scale; Step F2, use business knowledge and / or legal knowledge to modify the initial modification suggestion to obtain the modification suggestion.

[0054] Since the risk scale shows the acceptable modification scale for the review point, the text corresponding to the target review point can be compared with the risk scale hit by the target review point, and the initial modification suggestion can be determined based on the deviation degree between the two. To further avoid the modification suggestion exceeding the scope allowed by the business knowledge and / or legal knowledge matching the target business type, the business knowledge and / or legal knowledge matching the target business type can be used to modify the initial modification suggestion, so as to obtain a modification suggestion that not only lies within the acceptable modification scale range of the target review point but also conforms to the business knowledge and / or legal knowledge matching the target business type.

[0055] Table 1 shows the structure of a baseline. As shown in Table 1, in addition to the review point list, the baseline can also include business knowledge and / or legal knowledge.

[0056] Table 1

[0057] In the above process, the initial modification suggestion is first determined based on the deviation degree between the text corresponding to the target review point and the risk scale, and then the initial modification suggestion is modified using business knowledge and / or legal knowledge. Since business knowledge and / or legal knowledge are used as a backup, the modification suggestion finally obtained by this process has high guiding significance.

[0058] Due to different writing stances of the text, different interests are represented, and the corresponding text content and terms are usually different. For example, in a business cooperation contract, Party A, usually as the project leader or resource provider, will focus more on clarifying requirements such as Party B's service standards, delivery results, delivery time, etc.; Party B, as the service or product provider, is more concerned about its own rights and interests protection and reasonable profit space. The difference in text content will also make the corresponding baseline of the text different. In one implementation, step S104 can be performed as the following steps G1 - G2: Step G1, determine the target writing stance to which the target text belongs from the preset writing stances; Step G2, obtain the baseline matching the target business type and the target writing stance.

[0059] Among them, the preset writing stances can be at least two preset writing stances determined in advance for the target text. In addition to Party A and Party B in Table 1, the preset writing stances may also include other multiple parties such as Party C and Party D. Among them, Party C and Party D can be guarantors, witnesses, etc. The quantity and content included in the preset writing stances in this specification are not specifically limited and can be determined according to actual situations.

[0060] Specifically, the writing stance to which the target text belongs (i.e., the target writing stance) can be first determined from the preset writing stances. Furthermore, the target writing stance and the target business type can be combined to jointly determine a baseline, which not only matches the target business type but also matches the target writing stance. It should be noted that the baselines corresponding to different writing stances can be different, and the risk scales in these baselines can also be different.

[0061] It can be understood that, without violating the principle logic, the above-mentioned various method embodiments mentioned in this specification can be combined with each other to form combined embodiments. Due to space limitations, they are not elaborated in this specification. Those skilled in the art can understand that in the above methods of the specific implementation manner, the specific execution order of each step should be determined according to its function and possible internal logic.

[0062] It should be noted that for the method of text review provided in the embodiments of this specification, the execution subject can be a text review device, or a control module in the text review device for executing the method of text review. In the embodiments of this specification, the case where the text review device executes the method of text review is taken as an example to illustrate the text review device provided in the embodiments of this specification.

[0063] Figure 4 is a schematic structural diagram of a text review device according to an embodiment of the present invention. As Figure 4 shown, the text review device 400 includes: A business type module 410, configured to determine the target business type to which the target text belongs from multiple business types when receiving the target text to be subject to risk review; A baseline module 420, configured to obtain a baseline for discovering risks that matches the target business type, and the baseline at least includes a list of review points for characterizing potential risk points; A risk scale module 430, configured to determine the target review points hit by the target text from the list of review points, and obtain the risk scale for characterizing the acceptable risk deviation degree corresponding to the target review points; A first review module 440, configured to perform a first risk review on the target text based on the risk scale corresponding to the target review points to obtain a first risk review result.

[0064] In one embodiment, the target text is a contract text, and the risk scale module 430 includes: A text source unit for determining the target text source to which the target text belongs from a preset text source, where the preset text source includes standard contracts and non-standard contracts; A first key point unit for determining the target review key points hit by the target text from a review key point list based on the target text source.

[0065] In one embodiment, determining the target review key points hit by the target text from a review key point list based on the target text source includes: When the target text source is a standard contract, obtaining the target standard contract corresponding to the target business type; Comparing the target text with the target standard contract to determine the modification points; Based on the modification points, determining the target review key points hit by the target text from a review key point list.

[0066] In one embodiment, determining the target review key points hit by the target text from a review key point list based on the target text source includes: When the target text source is a non-standard contract, splitting the target text to obtain multiple sub-target texts; Determining the target review key points hit by the sub-target texts from a baseline; Merging the target review key points hit by each sub-target text included in the target text to obtain the target review key points hit by the target text.

[0067] In one embodiment, when the target text source is a non-standard contract, splitting the target text to obtain multiple sub-target texts includes: Obtaining the target title level used when splitting the target text; When the target text source is a non-standard contract, using the target title level to split the target text to obtain multiple sub-target texts.

[0068] In one embodiment, the text review device 400 further includes: A supplement module for obtaining business knowledge and / or legal knowledge matching the target business type from a baseline; A second review module for using the business knowledge and / or legal knowledge to perform a second risk review on the target text to obtain a second risk review result.

[0069] In one embodiment, the text review device 400 further includes: An initial suggestion module for determining an initial modification suggestion based on the deviation degree between the text corresponding to the target review key point and the risk scale; A modification verification module for modifying the initial modification suggestion using business knowledge and / or legal knowledge to obtain a modification suggestion.

[0070] In one embodiment, the baseline module 420 includes: A writing stance unit for determining the target writing stance to which the target text belongs from the preset writing stances; A baseline unit for obtaining a baseline that matches the target business type and the target writing stance.

[0071] It should be noted that the embodiments of the text review device in this specification and the embodiments of the text review method in this specification are based on the same inventive concept. Therefore, for the specific implementation of this embodiment, reference can be made to the corresponding text review method described above, and the repeated parts will not be elaborated.

[0072] Each module in the above text review device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the terminal device or the processor in the server in the form of hardware or be independent of them, or can be stored in the memory in the terminal device or the memory in the server in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0073] Furthermore, corresponding to the text review method described above, based on the same technical concept, one or more embodiments of this specification also provide an electronic device, which is used to execute the above text review method. Figure 5 It is a schematic structural diagram of an electronic device provided by one or more embodiments of this specification.

[0074] Based on the same idea, one or more embodiments of this specification also provide an electronic device, such as Figure 5As shown. Electronic devices can vary significantly due to differences in configuration or performance. They can include one or more processors 501 and a memory 502. The memory 502 can store one or more stored application programs or data. Among them, the memory 502 can be short-term storage or persistent storage. The application programs stored in the memory 502 can include one or more modules (not shown in the figure), and each module can include a series of computer-executable instructions for the electronic device. Further, the processor 501 can be configured to communicate with the memory 502 and execute a series of computer-executable instructions in the memory 502 on the electronic device. The electronic device can also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, and one or more keyboards 506.

[0075] In a specific embodiment, the electronic device includes a memory and one or more programs. One or more of the programs are stored in the memory, and one or more of the programs can include one or more modules. Each module can include a series of computer-executable instructions for the electronic device and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following: In the case of receiving a target text to be subject to risk review, determine the target business type to which the target text belongs from multiple business types; Obtain a baseline for risk discovery that matches the target business type. The baseline at least includes a list of review points for characterizing potential risk points; From the list of review points, determine the target review points hit by the target text, and obtain the risk scale corresponding to the target review points for characterizing the acceptable degree of risk deviation; Based on the risk scale corresponding to the target review points, perform a first risk review on the target text to obtain a first risk review result.

[0076] It should be noted that the embodiments of the electronic device in this specification and the embodiments of the method for text review in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the corresponding implementation of the text review method described above, and the repeated parts will not be elaborated.

[0077] Further, corresponding to the text review method described above, based on the same technical concept, one or more embodiments of this specification also provide a storage medium for storing computer-executable instructions. In a specific embodiment, the storage medium can be a USB flash drive, an optical disc, a hard disk, etc. When the computer-executable instructions stored in the storage medium are executed by a processor, the following process can be implemented: In the case of receiving a target text to be subject to risk review, determine the target business type to which the target text belongs from multiple business types; Obtain a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; From the list of review points, determine the target review points hit by the target text, and obtain the risk scale corresponding to the target review points for characterizing the acceptable degree of risk deviation; Based on the risk scale corresponding to the target review points, conduct a first risk review on the target text to obtain a first risk review result.

[0078] It should be noted that the embodiments of the storage medium in this specification and the method of text review in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the corresponding text review method described above, and the repeated parts will not be elaborated.

[0079] Furthermore, corresponding to the text review method described above, based on the same technical concept, one or more embodiments of this specification also provide a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, it can implement the following process: In the case of receiving a target text to be subject to risk review, determine the target business type to which the target text belongs from multiple business types; Obtain a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; From the list of review points, determine the target review points hit by the target text, and obtain the risk scale corresponding to the target review points for characterizing the acceptable degree of risk deviation; Based on the risk scale corresponding to the target review points, conduct a first risk review on the target text to obtain a first risk review result.

[0080] It should be noted that the embodiments of the computer program product in this specification and the embodiments of the text review method in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the corresponding text review method described above, and the repeated parts will not be elaborated.

[0081] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0082] In the 1990s, it was obvious to distinguish whether an improvement to a technology was a hardware improvement (e.g., improvement to circuit structures such as diodes, transistors, switches, etc.) or a software improvement (improvement to method flows). However, with the development of technology, many method flow improvements today can be regarded as direct improvements to hardware circuit structures. Almost all designers obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement to a method flow cannot be implemented with a hardware entity module. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program by themselves to "integrate" a digital system on a piece of PLD without having to ask a chip manufacturer to design and fabricate a dedicated integrated circuit chip. Moreover, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used in program development and writing. The original code before compilation also has to be written in a specific programming language, which is called a Hardware Description Language (HDL), and there is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones currently are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also be aware that by simply making a little logical programming of the method flow with the above-mentioned several hardware description languages and programming it into the integrated circuit, it is easy to obtain the hardware circuit that implements the logical method flow.

[0083] The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of the controller include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in the form of pure computer-readable program code, it is entirely possible to make the controller implement the same function in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or the structures within the hardware component.

[0084] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0085] For the convenience of description, the above devices are described by dividing them into various units according to their functions. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0086] Those skilled in the art should understand that one or more embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, one or more embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0087] This specification is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable text review devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable text review devices produce means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 one or more of the blocks for implementing the specified functions.

[0088] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable text review device to operate in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 one or more of the blocks for implementing the specified functions.

[0089] These computer program instructions can also be loaded onto a computer or other programmable text review device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 one or more of the blocks for implementing the specified functions.

[0090] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0091] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0092] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0093] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0094] One or more embodiments of this specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. One or more embodiments of this specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0095] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0096] The above are only examples of this document and are not intended to limit this document. For those skilled in the art, various changes and modifications can be made to this document. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this document shall be included within the scope of the claims of this document.

Claims

1. A method for text review, comprising: When receiving a target text to be subject to risk review, determining a target business type to which the target text belongs from multiple business types; Obtaining a baseline for risk discovery that matches the target business type, where the baseline at least includes a list of review points for characterizing potential risk points; Determining target review points hit by the target text from the list of review points, and obtaining a risk scale corresponding to the target review points for characterizing an acceptable degree of risk deviation; Based on the risk scale corresponding to the target review points, performing a first risk review on the target text to obtain a first risk review result.

2. The method according to claim 1, where the target text is a contract text, and determining target review points hit by the target text from the list of review points includes: Determining a target text source to which the target text belongs from a preset text source, where the preset text source includes standard contracts and non-standard contracts; Based on the target text source, determining target review points hit by the target text from the list of review points.

3. The method according to claim 2, where determining target review points hit by the target text from the list of review points based on the target text source includes: When the target text source is a standard contract, obtaining a target standard contract corresponding to the target business type; Comparing the target text with the target standard contract to determine modification points; Based on the modification points, determining target review points hit by the target text from the list of review points.

4. The method according to claim 2, where determining target review points hit by the target text from the list of review points based on the target text source includes: When the target text source is a non-standard contract, performing a splitting process on the target text to obtain multiple sub-target texts; Determining target review points hit by the sub-target texts from the baseline; Performing a merging process on the target review points hit by each of the sub-target texts included in the target text to obtain target review points hit by the target text.

5. The method according to claim 4, where performing a splitting process on the target text to obtain multiple sub-target texts when the target text source is a non-standard contract includes: Obtaining a target title level used for splitting the target text; When the target text source is a non-standard contract, using the target title level to perform a splitting process on the target text to obtain multiple sub-target texts.

6. The method according to claim 2, the method further includes: Obtaining business knowledge and / or legal knowledge that matches the target business type from the baseline; Using the business knowledge and / or the legal knowledge to perform a second risk review on the target text to obtain a second risk review result.

7. The method according to claim 6, the method further includes: Determine an initial modification suggestion based on the deviation degree between the text corresponding to the target review key point and the risk scale; Modify the initial modification suggestion by using the business knowledge and / or the legal knowledge to obtain a modification suggestion.

8. The method according to claim 1, wherein the obtaining of the baseline for risk discovery matching the target business type comprises: Determine a target writing stance to which the target text belongs from a preset writing stance; Obtain a baseline matching the target business type and the target writing stance.

9. A text review device, comprising: A business type module, configured to determine a target business type to which the target text belongs from multiple business types when receiving a target text to be subject to risk review; A baseline module, configured to obtain a baseline for risk discovery matching the target business type, where the baseline at least includes a list of review key points for characterizing potential risk points; A risk scale module, configured to determine a target review key point hit by the target text from the list of review key points, and obtain a risk scale corresponding to the target review key point for characterizing an acceptable risk deviation degree; A first review module, configured to perform a first risk review on the target text based on the risk scale corresponding to the target review key point to obtain a first risk review result.

10. An electronic device, comprising: A processor, and A memory arranged to store computer-executable instructions, which when executed, can cause the processor to: Determine a target business type to which the target text belongs from multiple business types when receiving a target text to be subject to risk review; Obtain a baseline for risk discovery matching the target business type, where the baseline at least includes a list of review key points for characterizing potential risk points; Determine a target review key point hit by the target text from the list of review key points, and obtain a risk scale corresponding to the target review key point for characterizing an acceptable risk deviation degree; Perform a first risk review on the target text based on the risk scale corresponding to the target review key point to obtain a first risk review result.

Images