Content auditing method and device, equipment, medium and program product

Abstract

The invention provides a content auditing method which can be applied to the technical field of artificial intelligence. The content auditing method comprises the steps of obtaining a to-be-audited text; based on the to-be-audited text, through preset retrieval constraints, audit regulations related to the to-be-audited text are retrieved from a preset knowledge vector database, the number of the preset retrieval constraints is N, and N is a positive integer; and executing content auditing on the to-be-audited text based on the auditing regulations to obtain an auditing result. The invention further provides a content auditing device and equipment, a storage medium and a program product.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, in particular to a content auditing method and device, equipment, medium and program product. BACKGROUND

[0002] The bank's product recommendation service based on SMS has the advantages of efficient reach, accuracy, low cost and instantaneity. It is not only a direct tool for promoting products and promoting product conversion, but also an effective channel for maintaining customer relationships, delivering important information and improving brand image, but it must be carried out under the premise of legality and compliance.

[0003] Before SMS pushing, how to ensure that the product promotion content is legal and compliant, reduce the differences of individual manual auditing and improve the auditing efficiency, has become an urgent need in the daily operation of commercial banks. In the prior art, a scheme of sensitive word matching is often used to realize the auditing of SMS pushing. However, only based on this scheme, all sensitive information cannot be covered, and there will still be certain risks. Therefore, how to ensure the accuracy and efficiency of the content checking of SMS pushing is a technical problem to be solved. SUMMARY

[0004] In view of the above problems, the present application provides a content auditing method and device, equipment, medium and program product which improve accuracy and efficiency.

[0005] According to a first aspect of the present application, a content auditing method is provided, comprising: obtaining a text to be audited; retrieving an auditing regulation related to the text to be audited from a preset knowledge vector database based on the text to be audited through a preset retrieval constraint, wherein the preset retrieval constraint includes N, and N is a positive integer; and performing content auditing on the text to be audited based on the auditing regulation to obtain an auditing result.

[0006] According to an embodiment of the present application, the establishment method of the preset knowledge vector database comprises: obtaining an auditing rule text; performing word segmentation processing on the auditing rule text to obtain L auditing regulations, L being a positive integer; performing classification based on the L auditing regulations to obtain P topics, P being a positive integer; and storing the L auditing regulations under the P topics respectively, wherein the auditing regulation and the topic are in a one-to-one or one-to-many relationship.

[0007] According to an embodiment of the present application, the classification based on the L auditing regulations to obtain P topics comprises: performing topic clustering of the L auditing regulations by a clustering algorithm.

[0008] In an embodiment of the present application, the preset retrieval condition includes Q theme constraints, Q being a positive integer, and retrieving the auditing regulation related to the text to be audited from the preset knowledge vector database based on the preset retrieval condition includes retrieving K auditing regulations related to the text to be audited from the preset knowledge vector database based on the text to be audited, K being a positive integer, and selecting W auditing regulations from the K auditing regulations based on the Q theme constraints, W being a positive integer.

[0009] In an embodiment of the present application, after the W auditing regulations are selected from the K auditing regulations based on the Q theme constraints, the method further includes obtaining the publication dates of the W auditing regulations, and selecting R auditing regulations from the W auditing regulations based on the publication dates.

[0010] In an embodiment of the present application, the content auditing based on the auditing regulation on the text to be audited to obtain an auditing result includes obtaining a prompt word, and outputting the auditing result based on the prompt word, the auditing regulation and the text to be audited as inputs of a preset machine learning model.

[0011] A second aspect of the present application provides a content auditing device, the device including: an obtaining module configured to obtain a text to be audited; a retrieving module configured to retrieve an auditing regulation related to the text to be audited from a preset knowledge vector database based on a preset retrieval constraint of the text to be audited, the preset retrieval constraint including N pieces, N being a positive integer; and an auditing module configured to perform content auditing on the text to be audited based on the auditing regulation to obtain an auditing result.

[0012] In an embodiment of the present application, the device further includes a knowledge base establishing module configured to obtain an auditing rule text, perform word segmentation processing on the auditing rule text to obtain L auditing regulations, L being a positive integer, perform classification based on the L auditing regulations to obtain P themes, P being a positive integer, and store the L auditing regulations under the P themes respectively, wherein the auditing regulation and the theme are in a one-to-one or one-to-many relationship.

[0013] In an embodiment of the present application, the knowledge base establishing module is specifically configured to perform theme clustering on the L auditing regulations by a clustering algorithm.

[0014] According to an embodiment of the present application, the preset retrieval condition comprises Q theme constraints, Q being a positive integer, the retrieval module is specifically configured to retrieve K auditing regulations related to the text to be audited from a preset knowledge vector database based on the text to be audited, K being a positive integer; and filter W auditing regulations from the K auditing regulations based on the Q theme constraints, W being a positive integer.

[0015] According to an embodiment of the present application, the retrieval module is further specifically configured to obtain the publication dates of the W auditing regulations; and filter R auditing regulations from the W auditing regulations based on the publication dates.

[0016] According to an embodiment of the present application, the auditing module is specifically configured to obtain a prompt word; and output the auditing result based on the prompt word, the auditing regulation and the text to be audited as inputs of a preset machine learning model.

[0017] A third aspect of the present application provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0018] A fourth aspect of the present application also provides a computer-readable storage medium having a computer program or instructions stored thereon, wherein the computer program or instructions are executed by a processor to implement the steps of the above method.

[0019] A fifth aspect of the present application also provides a computer program product comprising a computer program or instructions, wherein the computer program or instructions are executed by a processor to implement the steps of the above method.

[0020] To solve the technical problems of inaccurate and low-efficiency content auditing, embodiments of the present application retrieve relevant auditing regulations from a knowledge vector database based on the obtained auditing text and in cooperation with retrieval constraints, and audit the auditing text by the retrieved auditing regulations. Compared with the traditional content auditing by artificial retrieval and matching tools based on text, the embodiments of the present application can greatly improve the timeliness of auditing work, reduce the cost of artificial auditing, ensure the professionalism and consistency of the auditing result, and ensure that the product promotion work is carried out in accordance with the law. BRIEF DESCRIPTION OF DRAWINGS

[0021] The above content and other purposes, features and advantages of the present application will be more apparent from the following description of the embodiments of the present application with reference to the accompanying drawings, in which:

[0022] Figure 1 The application scenario diagram of the content auditing method, device, equipment, medium and program product according to an embodiment of the present application is schematically shown.

[0023] Figure 2 A flowchart of a content review method according to an embodiment of the present application is schematically shown;

[0024] Figure 3 A structural block diagram of a content review apparatus according to an embodiment of the present application is schematically shown; and

[0025] Figure 4 A block diagram of an electronic device suitable for implementing a content review method according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. It should be understood, however, that the description which follows is merely illustrative and is not intended to limit the scope of the present application. In the following detailed description of embodiments of the present application, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that one or more embodiments of the present application can be practiced without these specific details. In other instances, well-known structures and functions have not been described in detail in order to avoid obscuring aspects of the present application.

[0027] The terms used herein are merely used to describe specific embodiments and are not intended to limit the present application. The terms "include", "comprise", and the like used herein indicate the presence of the described features, steps, operations, and / or components but do not preclude the presence or addition of one or more other features, steps, operations, or components.

[0028] All terms used herein, including technical and scientific terms, have the same meanings as those generally understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having meanings that are consistent with the context of the present description, and should not be interpreted in an idealized or overly formal way.

[0029] In the case of using expressions similar to "at least one of A, B, and C, etc.", it should generally be interpreted to include at least one of each item enumerated, but not to exclude others not enumerated. For example, "a system having at least one of A, B, and C" should be interpreted to include a system having at least one of A, a system having at least one of B, a system having at least one of C, a system having at least one of A and B, a system having at least one of A and C, a system having at least one of B and C, and / or a system having at least one of A, B, and C, etc.

[0030] In a typical scenario, bank product promotion SMS as an efficient, accurate and low-cost customer touch tool plays a vital role in bank product and service promotion, promotion activities, and customer relationship management, but must meet regulatory compliance to avoid infringing privacy and harassing customers. Currently, the compliance audit of product promotion SMS is usually conducted by the bank compliance department according to the requirements of the Commercial Bank Financial Consumer Rights Protection Review Guidelines, the main laws and regulations related to the protection of the rights and interests of financial consumers and regulatory requirements, through text retrieval and matching tools, and the relevant provisions and inspection points are retrieved and matched one by one. The content of the product promotion SMS is manually audited for compliance.

[0031] The current bank product promotion SMS compliance audit process has the following shortcomings: first, the audit is fatigued and formalized, and the massive and high-frequency SMS demand may lead to compliance personnel auditing in form only, excessive reliance on keyword filtering (such as "highest", "guarantee"), and neglect of in-depth understanding of the overall context and the complexity of the preferential terms, there is a risk that the misleading risk is not fully identified. Second, the audit quality is highly dependent on individual ability and state, consistency is difficult to guarantee, and the understanding of the audit personnel of the complex financial regulations (advertising law, consumer protection law, individual protection law, special provisions for each financial product) and regulatory dynamics is different, the same SMS script may have different results when audited by different people at different times.

[0032] Embodiments of the present application provide a content auditing method, the method comprising: obtaining a text to be audited; retrieving, based on the text to be audited, an auditing regulation related to the text to be audited from a preset knowledge vector database through a preset retrieval constraint, wherein the preset retrieval constraint comprises N items, and N is a positive integer; and performing content auditing on the text to be audited based on the auditing regulation to obtain an auditing result.

[0033] To solve the technical problems of inaccurate and low-efficiency content auditing, embodiments of the present disclosure retrieve relevant auditing regulations from a knowledge vector library based on the obtained auditing text and in cooperation with retrieval constraints, and audit the auditing text based on the retrieved auditing regulations. Compared with the traditional manual auditing of content based on text retrieval and matching tools, the embodiments of the present disclosure can greatly improve the timeliness of auditing work, reduce the cost of manual auditing, ensure the professionalism and consistency of the auditing results, and ensure that the product promotion work is carried out in accordance with the law.

[0034] Figure 1 An application scenario diagram of the content auditing method according to an embodiment of the present application is schematically shown.

[0035] As Figure 1As shown, the application scenario 100 according to this embodiment can include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 can include various connection types, such as wired, wireless communication links, or fiber optic cables, and the like.

[0036] A user can use the first terminal device 101, the second terminal device 102, the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, and the like. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, and the like (only as examples).

[0037] The first terminal device 101, the second terminal device 102, the third terminal device 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, laptop computers, desktop computers, and the like.

[0038] The server 105 can be a server providing various services, such as a background management server supporting a website browsed by a user using the first terminal device 101, the second terminal device 102, the third terminal device 103 (only as an example). The background management server can analyze and process received user requests and the like, and feed back the processing results (such as a webpage, information, or data, and the like obtained or generated according to a user request) to the terminal device.

[0039] It should be noted that the content review method provided by the embodiments of the present application can generally be executed by the server 105. Correspondingly, the content review device provided by the embodiments of the present application can generally be arranged in the server 105. The content review method provided by the embodiments of the present application can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105. Correspondingly, the content review device provided by the embodiments of the present application can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105.

[0040] It should be understood that Figure 1 The number of terminal devices, networks, and servers in the application scenario 100 is only illustrative. According to the implementation needs, there can be any number of terminal devices, networks, and servers.

[0041] The following will be based on Figure 1 The described scenario, by Figure 2 The content review method according to the embodiments of the application is described in detail.

[0042] Figure 2 The flowchart of the content review method according to the embodiments of the application is schematically shown.

[0043] As Figure 2 The content review method of this embodiment includes operations S210-S230.

[0044] In operation S210, the text to be reviewed is obtained.

[0045] The text to be reviewed can be a text for product promotion, which can be a short message text or a long text with more details.

[0046] According to an embodiment of the present disclosure, the method for establishing the preset knowledge vector database comprises: obtaining a review rule text; performing word segmentation processing on the review rule text to obtain L review regulations, L being a positive integer; performing classification based on the L review regulations to obtain P topics, P being a positive integer; and storing the L review regulations under the P topics respectively, wherein the review regulations and the topics are in a one-to-one or one-to-many relationship.

[0047] The preset knowledge vector database is used to store various review regulations, which can be direct descriptions such as prohibiting the appearance of XX words, or indirect descriptions such as not appearing XX descriptions. In a typical scenario, the preset knowledge vector database is responsible for vectorizing and encoding documents such as commercial bank financial consumer rights protection review points guidelines, main laws and regulations related to financial consumer rights protection, and supervision regulations, and storing them in the knowledge vector database.

[0048] Specifically, before the detailed review of the text to be reviewed, a knowledge vector database for storing review regulations needs to be established in advance. The obtained review rule text is segmented to obtain L individual review regulations, and each review regulation is classified under a pre-set topic to establish the knowledge vector database. It should be noted that the same review regulation can be under multiple different topic classifications.

[0049] In a typical scenario, the documents of the commercial bank financial consumer rights protection review points guide, the main laws and regulations and regulatory provisions involved in the financial consumer rights protection, etc. are preprocessed, including removing special characters, removing irrelevant information, removing repetitive descriptions, word segmentation, etc. The word vector algorithm is used to extract the feature vector of the commercial bank financial consumer rights protection review points guide, the main laws and regulations and regulatory provisions involved in the financial consumer rights protection, etc. According to the 9 categories of topics of the right to know, the right to choose, the right to fair trade, the right to property safety, the right to be respected, the right to seek compensation, the right to information security, the right to education, and the right to privacy protection, the inspection rules of the same type of rights and interests are classified into a category.

[0050] In addition, the document knowledge block vector of the commercial bank financial consumer rights protection review points guide, the main laws and regulations and regulatory provisions involved in the financial consumer rights protection, etc. and the knowledge block metadata label (right type, release date as metadata label) are stored in the vector database together, and a retrieval index is established with the right type and metadata label as the query condition.

[0051] According to the embodiments of the present disclosure, the classification based on the L audit regulations is performed to obtain P topics, including: clustering the L audit regulation execution topics by a clustering algorithm.

[0052] Specifically, the L audit regulation execution topics are automatically clustered by means of a clustering algorithm. For example, the L audit regulations are automatically classified by a K-means algorithm.

[0053] In a typical scenario, the clustering algorithm is used to cluster the function menu by topic, and the commercial bank financial consumer rights protection review points guide, the main laws and regulations and regulatory provisions involved in the financial consumer rights protection, etc. are divided into knowledge blocks under different right types according to the clustering results.

[0054] In operation S220, based on the text to be audited, the audit regulations related to the text to be audited are retrieved from the preset knowledge vector database by a preset retrieval constraint, wherein the preset retrieval constraint includes N, and N is a positive integer.

[0055] Among them, the preset retrieval constraint can be a constraint of a specific field, a constraint of a retrieval formula, or a semantic constraint, which is not limited here.

[0056] Specifically, the text to be audited is matched to the similar audit regulations by semantic matching, and then the range of the audit regulations is limited by the preset retrieval constraints to obtain the final audit regulations to be used.

[0057] According to an embodiment of the present disclosure, the preset retrieval condition comprises Q theme constraints, Q being a positive integer, and the retrieving, based on the to-be-audited text, the auditing regulations related to the to-be-audited text from the preset knowledge vector database comprises: retrieving, based on the to-be-audited text, K auditing regulations related to the to-be-audited text from the preset knowledge vector database, K being a positive integer; and selecting W auditing regulations from the K auditing regulations based on the Q theme constraints, W being a positive integer.

[0058] Specifically, Q themes can be selected as constraints according to the themes of the knowledge vector database, so as to limit the retrieval range of the auditing regulations in the knowledge vector database, and output W auditing regulations for use in the subsequent operation S230.

[0059] In a typical scenario, the product promotion message content is used to extract a feature vector, which is converted into a product promotion message vector representation V1. A vector database query is initiated, and the retrieval conditions Q1...Q9 are informed right, self-selection right, fair trade right, property safety right, respect right, legal compensation right, information security right, education right, and privacy protection right. All document knowledge block vectors that meet the retrieval conditions Qi are retrieved.

[0060] According to an embodiment of the present disclosure, after the W auditing regulations are selected from the K auditing regulations based on the Q theme constraints, the method further comprises: obtaining the publication dates of the W auditing regulations; and selecting R auditing regulations from the W auditing regulations based on the publication dates.

[0061] Specifically, on the basis of the W auditing regulations, the latest R auditing regulations can also be selected according to the publication dates for use in the subsequent operation S230.

[0062] In a typical scenario, the latest knowledge block vector is obtained according to the order from recent to remote, and the text description K1 to Kn corresponding to the knowledge block vector is returned. The filtering algorithm based on the knowledge block publication date sorting is as follows: Kn=Sort(W,KV), wherein KV is all knowledge block vectors in the i th right classification, W is a weight calculated according to the publication date, and W=1 / (1+current date-publication date).

[0063] In operation S230, content auditing is performed on the to-be-audited text based on the auditing regulations, and an auditing result is obtained.

[0064] Specifically, the to-be-audited text is audited by checking the regulations, wherein, when the regulations are explicit regulations, the to-be-audited text is audited by automatic software to avoid the appearance of XX words; when the regulations are non-explicit regulations, the to-be-audited text can be identified by combining a machine learning model to ensure that the to-be-audited text complies with the corresponding audit regulations. The audit result includes passing the audit and failing the audit, and when the audit fails, the abnormal field can also be prompted for personnel to modify.

[0065] According to an embodiment of the present disclosure, the content audit is performed on the to-be-audited text based on the audit regulations to obtain an audit result, including: obtaining a prompt word; and based on the prompt word, the audit regulations, and the to-be-audited text as inputs of a preset machine learning model, outputting the audit result.

[0066] Specifically, the audit can also be further implemented by obtaining a prompt word to further realize personalized audit, and the prompt word, the audit regulations, and the to-be-audited text are inputted into a preset machine learning model to output an audit result limited by the prompt word, wherein the preset machine learning model can be obtained by fine-tuning an initial large language model, which will not be described here.

[0067] In a typical scenario, to realize compliance audit of product promotion scheme text, the system presets a corresponding consumer protection audit prompt word template, and completes automatic consumer protection clause audit by calling a large language model. The preset consumer protection audit prompt word template includes the following contents: consumer protection audit task description, consumer protection audit check process, product promotion scheme, output format requirement, etc. For example, the consumer protection audit prompt word template is set as follows: you are a consumer protection law audit expert of a commercial bank, familiar with the consumer protection audit process and rules for product promotion short message, please judge whether the content of the product promotion short message text complies with the consumer protection regulations, and screen out the product promotion short message that does not comply with the requirements. When auditing the product promotion scheme, please follow the following rules: 1. Audit the rights and interests according to the provisions of the consumer protection law. 2. Follow the steps of the commercial bank consumer protection audit points guide to check in the audit process to avoid omissions. 3. The content of the consumer protection rights and interests and the audit guide reference vector data retrieval {Kn}. 4. The product promotion short message text reference input {V1}. 5. Do not arbitrarily add or imagine content not in the consumer protection rights and interests and the audit guide in the detection. The output content is returned in JSON format, including whether the product promotion short message exists compliance, violation risk points, violation judgment reference clauses and regulations, and adjustment suggestions.

[0068] To solve the technical problems of inaccurate and low-efficiency content review, embodiments of the present disclosure retrieve relevant review regulations from a knowledge vector library based on the obtained review text and in cooperation with retrieval constraints, and review the review text based on the retrieved review regulations. Compared with the traditional content review by artificial text retrieval and matching tools, the embodiments of the present disclosure can greatly improve the timeliness of review work, reduce the cost of artificial review, ensure the professionalism and consistency of review results, and ensure that product promotion work is carried out in accordance with laws and regulations.

[0069] Based on the above content review method, the present application also provides a content review device. The following will be described in detail Figure 3 The device is described in detail.

[0070] Figure 3 The structure block diagram of the content review device according to the embodiments of the present application is schematically shown.

[0071] As Figure 3 shown, the content review device 300 of the embodiments includes an acquisition module 310, a retrieval module 320, and a review module 330.

[0072] The acquisition module 310 is configured to acquire a review text. In an embodiment, the acquisition module 310 can be configured to perform the operation S210 described above, and details are not repeated here.

[0073] The retrieval module 320 is configured to retrieve, based on the review text, a review regulation related to the review text from a preset knowledge vector database through a preset retrieval constraint. The preset retrieval constraint includes N pieces, and N is a positive integer. In an embodiment, the retrieval module 320 can be configured to perform the operation S220 described above, and details are not repeated here.

[0074] The review module 330 is configured to perform content review on the review text based on the review regulation, and obtain a review result. In an embodiment, the review module 330 can be configured to perform the operation S230 described above, and details are not repeated here.

[0075] To solve the technical problems of inaccurate and low-efficiency content review, embodiments of the present disclosure retrieve relevant review regulations from a knowledge vector library based on the obtained review text and in cooperation with retrieval constraints, and review the review text based on the retrieved review regulations. Compared with the traditional content review by artificial text retrieval and matching tools, the embodiments of the present disclosure can greatly improve the timeliness of review work, reduce the cost of artificial review, ensure the professionalism and consistency of review results, and ensure that product promotion work is carried out in accordance with laws and regulations.

[0076] According to an embodiment of the present application, the device further comprises a knowledge base establishing module configured to: obtain an audit rule text; perform a word segmentation processing on the audit rule text to obtain L audit regulations, L being a positive integer; perform a classification based on the L audit regulations to obtain P topics, P being a positive integer; and store the L audit regulations under the P topics respectively, wherein the audit regulations and the topics are in a one-to-one or one-to-many relationship.

[0077] According to an embodiment of the present application, the knowledge base establishing module is specifically configured to perform a topic clustering on the L audit regulations by a clustering algorithm.

[0078] According to an embodiment of the present application, the preset retrieval condition comprises Q topic constraints, Q being a positive integer, and the retrieval module is specifically configured to: retrieve K audit regulations related to the text to be audited from a preset knowledge vector database based on the text to be audited, K being a positive integer; and filter W audit regulations from the K audit regulations based on the Q topic constraints, W being a positive integer.

[0079] According to an embodiment of the present application, the retrieval module is further specifically configured to: obtain a publication date of the W audit regulations; and filter R audit regulations from the W audit regulations based on the publication date.

[0080] According to an embodiment of the present application, the audit module is specifically configured to: obtain a prompt word; and output the audit result based on the prompt word, the audit regulations and the text to be audited as inputs of a preset machine learning model.

[0081] According to an embodiment of the present application, any of the modules of the obtaining module 310, the retrieval module 320 and the audit module 330 can be combined in one module, or any of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of the modules can be combined with at least part of the functions of other modules, and implemented in one module. According to an embodiment of the present application, at least one of the obtaining module 310, the retrieval module 320 and the audit module 330 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on board, a system on package, an application specific integrated circuit (ASIC), or any other reasonable manner of integrating or packaging a circuit, etc. or implemented by hardware or firmware, or implemented in any one of software, hardware and firmware or in a proper combination of any of them. Alternatively, at least one of the obtaining module 310, the retrieval module 320 and the audit module 330 can be at least partially implemented as a computer program module which can perform corresponding functions when the computer program module is run.

[0082] Figure 4 A block diagram of an electronic device suitable for implementing the content review method according to embodiments of the present application is shown schematically.

[0083] As shown in Figure 4 The electronic device 900 according to embodiments of the present application includes a processor 901 that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 902 or loaded from a storage section 908 into a random access memory (RAM) 903. The processor 901 can include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor, and / or a related chipset, and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), and so on. The processor 901 can also include an on-board memory for cache use. The processor 901 can include a single processing unit or multiple processing units for executing different actions of the method processes according to embodiments of the present application.

[0084] In the RAM 903, various programs and data required for the operation of the electronic device 900 are stored. The processor 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. The processor 901 performs various operations of the method processes according to embodiments of the present application by executing the programs in the ROM 902 and / or the RAM 903. Note that the programs can also be stored in one or more memories other than the ROM 902 and the RAM 903. The processor 901 can also perform various operations of the method processes according to embodiments of the present application by executing the programs stored in the one or more memories.

[0085] According to embodiments of the present application, the electronic device 900 can also include an input / output (I / O) interface 905 that is also connected to the bus 904. The electronic device 900 can also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as necessary. A removable medium 911 such as a magnetic disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 910 as necessary, so that a computer program read out from it is installed in the storage section 908 as necessary.

[0086] The application further provides a computer readable storage medium, which can be included in the device / apparatus / system described in the above embodiments, or can exist independently without being assembled into the device / apparatus / system. The computer readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the application.

[0087] According to the embodiments of the application, the computer readable storage medium can be a non-volatile computer readable storage medium, which can include, but is not limited to, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In this application, a computer readable storage medium can be any tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to the embodiments of the application, the computer readable storage medium can include one or more of the above-described ROM 902 and / or RAM 903 and / or one or more memories other than the ROM 902 and the RAM 903.

[0088] The embodiments of the application also include a computer program product, which includes a computer program containing program codes for executing the methods shown in the flowcharts. When the computer program product is run in a computer system, the program codes are used to make the computer system implement the methods provided by the embodiments of the application.

[0089] The above-described functions defined in the system / apparatus of the embodiments of the application are performed when the computer program is executed by the processor 901. According to the embodiments of the application, the above-described system, apparatus, module, unit, etc. can be implemented by computer program modules.

[0090] In one embodiment, the computer program can rely on a tangible storage medium such as an optical storage medium, a magnetic storage medium, etc. In another embodiment, the computer program can also be transmitted, distributed, and downloaded in the form of a signal on a network medium, and be downloaded and installed through the communication part 909, and / or installed from the detachable medium 911. The program codes contained in the computer program can be transmitted by any appropriate network medium, including but not limited to wireless, wired, etc., or any appropriate combination thereof.

[0091] In such embodiments, the computer program can be downloaded and installed from the network via the communication section 909, and / or installed from the removable media 911. When the computer program is executed by the processor 901, the above-described functions defined in the system of the embodiments of the present application are performed. According to the embodiments of the present application, the system, device, apparatus, module, unit, and the like described above can be implemented by the computer program modules.

[0092] According to the embodiments of the present application, the program code for executing the computer program provided by the embodiments of the present application can be written in any combination of one or more programming languages, and specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming language, and / or assembly / machine language. The programming language includes, but is not limited to, such as Java, C++, python, "C" language, or similar programming language. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case involving a remote computing device, the remote computing device can be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, connected to the Internet through an Internet service provider).

[0093] The flowcharts and block diagrams in the drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a portion of code, which contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that shown in the figures. For example, two blocks noted in succession can actually be executed substantially concurrently, or they can sometimes be executed in reverse order, depending on the functionality involved. It should also be noted that each block in the flowcharts or block diagrams, and combinations of blocks in the flowcharts or block diagrams, can be implemented by dedicated hardware-based systems that perform the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0094] Those skilled in the art can understand that the features described in various embodiments of the present application can be combined and / or integrated in various combinations, even if such combinations are not explicitly described in the present application. In particular, the features described in various embodiments of the present application can be combined and / or integrated in various combinations without departing from the spirit and teachings of the present application. All such combinations and / or integrations are within the scope of the present application.

Claims

1. A content review method characterized by, The method comprises: acquiring a text to be audited; retrieving, based on the text to be audited, auditing regulations related to the text to be audited from a preset knowledge vector database through a preset retrieval constraint, wherein the preset retrieval constraint comprises N pieces, and N is a positive integer; and performing content auditing on the text to be audited based on the auditing regulations to obtain an auditing result.

2. The method of claim 1, wherein, The method for establishing the preset knowledge vector database comprises: acquiring an auditing rule text; performing word segmentation processing on the auditing rule text to obtain L auditing regulations, and L is a positive integer; performing classification based on the L auditing regulations to obtain P topics, and P is a positive integer; and storing the L auditing regulations under the P topics respectively, wherein the auditing regulations and the topics are in a one-to-one or one-to-many relationship.

3. The method of claim 2, wherein performing classification based on the L auditing regulations to obtain P topics comprises: performing topic clustering on the L auditing regulations through a clustering algorithm.

4. The method of claim 2, wherein, The preset retrieval condition comprises Q topic constraints, and Q is a positive integer. The method for retrieving, based on the text to be audited, auditing regulations related to the text to be audited from a preset knowledge vector database through a preset retrieval condition comprises: retrieving, based on the text to be audited, K auditing regulations related to the text to be audited from a preset knowledge vector database, and K is a positive integer; and filtering W auditing regulations from the K auditing regulations based on the Q topic constraints, and W is a positive integer.

5. The method of claim 4, wherein after filtering W auditing regulations from the K auditing regulations based on the Q topic constraints, the method further comprises: acquiring publication dates of the W auditing regulations; and filtering R auditing regulations from the W auditing regulations based on the publication dates.

6. The method of any one of claims 1-5, wherein performing content auditing on the text to be audited based on the auditing regulations to obtain an auditing result comprises: acquiring prompt words; and outputting the auditing result based on the prompt words, the auditing regulations, and the text to be audited as inputs of a preset machine learning model. The device comprises: an acquisition module configured to acquire a text to be audited; 7. A content review apparatus characterized by comprising: a retrieval module configured to retrieve, based on the text to be audited, auditing regulations related to the text to be audited from a preset knowledge vector database through a preset retrieval constraint, wherein the preset retrieval constraint comprises N pieces, and N is a positive integer; and an auditing module configured to perform content auditing on the text to be audited based on the auditing regulations to obtain an auditing result.

8. An electronic device comprising: one or more processors; a memory configured to store one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-6. The computer program or instructions, when executed by a processor, implement the steps of the method according to any one of claims 1-6. ​ 9. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, ​ 10. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instructions implement the steps of the method according to any one of claims 1-6 when executed by a processor.

Images