Data compliance auditing method and device, equipment, storage medium and program product

Abstract

The invention discloses a data compliance auditing method and device, equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining enterprise WeChat data, and classifying the enterprise WeChat data to obtain different types of enterprise WeChat data; performing data compliance auditing on different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain a preliminary auditing result; and performing iterative optimization on a pre-trained natural language processing model and a pre-trained computer vision model based on the preliminary audit result. According to the scheme, compliance auditing is carried out on different types of enterprise WeChat data based on the natural language processing model and the computer vision model, the auditing efficiency and accuracy can be effectively improved, and the cost of manual intervention is reduced. And meanwhile, iterative optimization of the model is performed according to a preliminary auditing result, so that the auditing accuracy is further improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to data compliance audit methods, devices, equipment, storage media, and program products. Background Art

[0002] With the rapid development of the Internet and the widespread use of WeChat for Business, a large amount of data has been generated in WeChat for Business, including text, pictures, videos and other forms. Some of these data may contain illegal content, such as sensitive words and bad information, which require compliance review. However, traditional data review methods mainly rely on manual or keyword recognition, which is inefficient and prone to errors.

[0003] In the current market, a series of enterprise WeChat monitoring solutions have emerged, but their functional diversity still needs to be improved. These tools generally focus on text analysis and lack in-depth verification and evaluation mechanisms for multimedia data, which undoubtedly leaves blind spots when dealing with increasingly rich digital interactions. In terms of the covered enterprise WeChat business scenarios, different tools appear fragmented, some are limited to monitoring chat conversations, while others are limited to tracking the dynamics of the circle of friends, lacking comprehensive coverage capabilities. In addition, the data detection strategies of these tools appear to be immature, and the accuracy of identifying potential risks is insufficient, resulting in a high false alarm rate. This may not only affect the company's capture of key information, but may also cause unnecessary troubles and waste of resources in actual operations. Therefore, for an ideal enterprise WeChat monitoring tool, it is crucial to improve the accuracy of risk identification, establish a more intelligent and comprehensive monitoring framework, and optimize support for various data types. Summary of the invention

[0004] The main purpose of this application is to provide a data compliance audit method, device, equipment, storage medium and program product, aiming to solve the technical problem of how to efficiently and accurately perform compliance audits on various types of data in enterprise WeChat.

[0005] To achieve the above objectives, this application proposes a data compliance audit method, which includes: Acquire enterprise WeChat data, and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; Perform data compliance review on the different types of enterprise WeChat data by using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results; Based on the preliminary review results, the pre-trained natural language processing model and the pre-trained computer vision model are iteratively optimized.

[0006] In one embodiment, the step of performing data compliance review on the different types of enterprise WeChat data by using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results includes: For different types of enterprise WeChat data, select the corresponding pre-trained natural language processing model and / or pre-trained computer vision model to conduct data compliance review and obtain preliminary review results.

[0007] In one embodiment, the different types of enterprise WeChat data include text data, pictures and video data. For the different types of enterprise WeChat data, the corresponding pre-trained natural language processing model and / or pre-trained computer vision model are selected to perform data compliance review. The step of obtaining the preliminary review result includes: Processing text data using a pre-trained natural language processing model to extract illegal text content; and / or Process image and video data using pre-trained computer vision models to identify illegal visual content; The illegal text content and the illegal visual content are summarized to generate a preliminary review result.

[0008] In one embodiment, before the step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review result, the step further includes: Send the preliminary review result to the user, so that the user can verify it according to the preliminary review result and return the verification result; The verification result returned by the user is obtained, and the verification result is sent to a pre-trained natural language processing model and a pre-trained computer vision model.

[0009] In one embodiment, the step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review result includes: Based on the verification results returned by the users, the pre-trained natural language processing model and the pre-trained computer vision model are iteratively optimized.

[0010] In one embodiment, the step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model according to the verification result returned by the user includes: Confirm new illegal content or new appeal results based on the verification results returned by the user; In a pre-trained natural language processing model and a pre-trained computer vision model, iterative optimization is performed based on the new illegal content or the new complaint result.

[0011] In addition, to achieve the above purpose, the present application also proposes a data compliance audit device, the data compliance audit device comprising: A classification module is used to obtain enterprise WeChat data and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; An audit module, used to perform data compliance audit on the different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary audit results; An optimization module is used to iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review results.

[0012] In addition, to achieve the above-mentioned purpose, the present application also proposes a data compliance audit device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the data compliance audit method described above.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the data compliance audit method described above are implemented.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the data compliance audit method described above.

[0015] The embodiments of the present application provide a data compliance audit method, device, equipment, storage medium and program product, which specifically obtains different types of enterprise WeChat data by acquiring enterprise WeChat data and classifying the enterprise WeChat data; performs data compliance audit on different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary audit results; based on the preliminary audit results, iteratively optimizes the pre-trained natural language processing model and the pre-trained computer vision model. This solution performs compliance audits on different types of enterprise WeChat data based on natural language processing models and computer vision models, which can effectively improve audit efficiency and accuracy and reduce the cost of manual intervention. At the same time, iterative optimization of the model is performed based on the preliminary audit results to further improve the audit accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 A flowchart of the first embodiment of the application data compliance review method is provided; Figure 2 A flowchart of the second embodiment of the data compliance review method of this application; Figure 3 A brief flowchart of the data compliance review method provided in Example 1 of the present application; Figure 4 This is a schematic diagram of the module structure of the data compliance audit device according to an embodiment of the present application; Figure 5 Schematic diagram of the device structure of the hardware operating environment involved in the data compliance audit method in the embodiment of the present application.

[0019] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0021] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0022] The main solution of the embodiments of the present application is: to conduct data compliance review on different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results; based on the preliminary review results, to iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model.

[0023] In the current market, a series of enterprise WeChat monitoring solutions have emerged, but their functional diversity still needs to be improved. These tools generally focus on text analysis and lack in-depth verification and evaluation mechanisms for multimedia data, which undoubtedly leaves blind spots when dealing with increasingly rich digital interactions. In terms of the covered enterprise WeChat business scenarios, different tools appear fragmented, some are limited to monitoring chat conversations, while others are limited to tracking the dynamics of the circle of friends, lacking comprehensive coverage capabilities. In addition, the data detection strategies of these tools appear to be immature, and the accuracy of identifying potential risks is insufficient, resulting in a high false alarm rate. This may not only affect the company's capture of key information, but may also cause unnecessary troubles and waste of resources in actual operations. Therefore, for an ideal enterprise WeChat monitoring tool, it is crucial to improve the accuracy of risk identification, establish a more intelligent and comprehensive monitoring framework, and optimize support for various data types.

[0024] This application provides a solution that applies advanced computer vision technology to analyze multimedia files containing pictures and videos in enterprise WeChat data, which expands the scope of traditional text review and improves the review system's ability to handle multimedia content. By combining natural language models and computer vision models, enterprise WeChat content review tasks can be automated, reducing the need for manual intervention, thereby improving processing speed and efficiency. In addition to fully covering the enterprise WeChat conversation content review function, this solution is further extended to the review of enterprise WeChat Moments and customer notes, ensuring that all interactions and information records meet strict compliance standards. In this way, companies can not only maintain transparency and professionalism in daily communication, but also meet regulatory requirements in the management of shared content in Moments and personalized notes for customers, thereby protecting the company's reputation and user data security in all aspects.

[0025] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, etc. The following takes a mobile phone as an example to illustrate this embodiment and the following embodiments.

[0026] Based on this, the present application embodiment provides a data compliance audit method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the application data compliance review method.

[0027] In this embodiment, the data compliance audit method includes steps S10 to S30: Step S10, obtaining enterprise WeChat data, and classifying the enterprise WeChat data to obtain different types of enterprise WeChat data; It should be noted that enterprise WeChat data includes but is not limited to conversation data, circle of friends data, and customer note data.

[0028] Since in these enterprise WeChat data, not only text, video or pictures may exist separately, but also one enterprise WeChat data may contain text and pictures, text and video or multiple contents at the same time, so step S10 can effectively classify the complex and diverse enterprise WeChat data, so as to facilitate the subsequent compliance review of the enterprise WeChat data.

[0029] Specifically, the API interface provided by WeChat for Business is called to obtain chat record information, circle of friends data or customer note data, etc., where these data include text data, image data, video data, etc. The obtained WeChat for Business data is stored in a local database or file system, or in the cloud, for subsequent data classification.

[0030] For text data, natural language processing technology is used to extract keywords or phrases from the text content, and then sentiment analysis tools are used to determine the emotional tendency of the text content, further identify the theme of the text, and then store the text data in document files.

[0031] For the classification of image and video data, images are classified by using computer vision technology, such as OCR recognition, object recognition, etc. For video data, key frames can be extracted for image analysis, or deep learning algorithms can be used for content recognition, and the image and video data are stored in media files. However, since video data contains not only video but also audio, it is also necessary to extract pure audio data and pure video data from the video data, and store the pure audio data in audio files.

[0032] Through the above steps, the enterprise WeChat data can be classified, which is conducive to the subsequent targeted data compliance review. In addition to comprehensively covering the chat record content of the enterprise WeChat, it is further expanded to the review of the enterprise WeChat Moments and customer notes, ensuring that all interactions and information records meet strict compliance standards. In this way, the company can not only maintain transparency and professionalism in daily communication, but also meet regulatory requirements in the sharing content of the Moments and the management of personalized notes for customers, thereby protecting the company's reputation and user data security in all aspects.

[0033] Step S20, performing data compliance review on the different types of enterprise WeChat data using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain a preliminary review result; It should be noted that the natural language processing model is a computer program and technology used to process and understand human language. The model can perform a variety of tasks, including but not limited to text classification, sentiment analysis, named entity recognition, machine translation, question-answering system, text generation, etc.

[0034] Computer vision models are algorithms and techniques used to process and understand image or video data. These models are able to extract useful information from images or videos and complete various tasks.

[0035] In a feasible implementation, step S20 may include step S21: Step S21, for different types of enterprise WeChat data, select the corresponding pre-trained natural language processing model and / or pre-trained computer vision model to perform data compliance review and obtain preliminary review results.

[0036] Optionally, in order to make the enterprise WeChat data compliance review system show higher accuracy and reliability, this embodiment uses a natural language processing model to process text information, and a computer vision model to process image and video data. On social platforms such as enterprise WeChat, text, pictures, and videos are often interrelated. For example, a user may post content that contains text descriptions and pictures or videos. The natural language model can analyze text content, and the computer vision model can analyze pictures and videos. The combination of the two can provide a more comprehensive content analysis.

[0037] Further, in a feasible implementation manner, step S21 may include steps A21 to A23: Step A21, processing the text data through a pre-trained natural language processing model to extract illegal text content; Specifically, based on the enterprise WeChat data classified in step S10, the text data, image data and video data are processed separately.

[0038] For text or document files, first use a pre-trained natural language processing model to remove irrelevant information in the text or document files, such as punctuation and special characters. Then the text in the text needs to be standardized, such as unifying uppercase and lowercase, abbreviations, etc., so that the model can understand it, and then break the sentence into separate words or phrases. Convert the words into numerical vectors so that the natural language processing model understands the meaning and context of the sentence and extracts the keywords in the text data. Next, use the extracted keywords to match potentially illegal words. You can also evaluate the sentiment in the text, detect negative emotions or malicious comments, or identify sensitive information such as personal names and organization names, and mark these potentially illegal words, negative emotions or malicious comments and sensitive information.

[0039] For each text segment marked as a violation, the specific violation type, violation content, context, and other information are recorded, and this information is summarized to generate and output the text content of the violation.

[0040] For audio files, the audio content needs to be converted into a text format that can be processed by the natural language processing model. Once the audio is converted into text, the natural language processing model is used to analyze the text and extract the corresponding illegal content.

[0041] Step A22, processing the image and video data using a pre-trained computer vision model to identify illegal visual content; For image and video data stored in multimedia files, it is necessary to convert the images and videos into a unified format and adjust the images to the size and brightness or contrast required by the computer vision model to obtain preprocessed images and videos. Then, the computer vision model identifies the objects in the image or video, confirms the scene depicted in the image or video, and extracts any visible text from the image. These steps can extract effective features from the image or video.

[0042] The extracted valid features are further analyzed and identified, such as whether these valid features contain sensitive objects, whether there is inappropriate behavior, and whether they contain illegal words.

[0043] These sensitive objects, inappropriate behaviors, and illegal words are summarized to obtain illegal visual content and use it as the output result.

[0044] Through the above steps, advanced computer vision models are applied to analyze multimedia files containing pictures and videos in enterprise micro data, which not only expands the scope of traditional text review, but also improves the review system's ability to handle multimedia content.

[0045] Step A23, summarizing the illegal text content and the illegal visual content to generate a preliminary review result.

[0046] All illegal contents identified in step A21 and step A22 are summarized to generate a preliminary review result document containing all illegal contents, including violations of text, images, videos and audio.

[0047] Through the above steps, combining natural language models and computer vision models can automate many enterprise micro-content review tasks, reduce the need for human intervention, and thus improve processing speed and efficiency.

[0048] Step S30, based on the preliminary review result, iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model.

[0049] In order to further improve the review accuracy of enterprise WeChat data, in the embodiment, it is also necessary to iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model.

[0050] In a feasible implementation, step S30 may include step S31: Step S31, iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model according to the verification result returned by the user.

[0051] It should be noted that the verification results refer to the feedback and appeal results returned by users after professional verification based on the preliminary review results.

[0052] Further, in a feasible implementation manner, step S31 may include steps B31-B32: Step B31, confirming new illegal content or new appeal results based on the verification results returned by the user; According to the feedback and appeal results returned by users, confirm which are the newly added illegal content or illegal words. According to the confirmed new illegal content or illegal status, update the illegal content list or remove the illegal status of certain content.

[0053] Step B32, performing iterative optimization in a pre-trained natural language processing model and a pre-trained computer vision model based on the new illegal content or the new complaint result.

[0054] The content confirmed as new violations is added to the training dataset for training the computer vision model; at the same time, the confirmed text complaint content is added to the dataset of the natural language processing model.

[0055] Input the updated violation data set into the natural language processing model and computer vision model, and repeat steps S10 to S20 to achieve iterative optimization of the natural language processing model and computer vision model, and improve the accuracy and generalization ability of the natural language processing model and computer vision model. After the training is completed, the model needs to be evaluated to ensure that its performance on the new data meets expectations. Finally, the optimized natural language processing model and computer vision model are deployed to the production environment to replace the old version of the natural language processing model and computer vision model.

[0056] The above-mentioned embodiment method includes obtaining enterprise WeChat data and classifying the enterprise WeChat data to obtain different types of enterprise WeChat data; performing data compliance audit on different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary audit results; and iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary audit results. This solution performs compliance audit on different types of enterprise WeChat data based on natural language processing models and computer vision models, which can effectively improve audit efficiency and accuracy and reduce the cost of manual intervention. At the same time, iterative optimization of the model is performed based on the preliminary audit results to further improve the audit accuracy.

[0057] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 Before step S30, the data compliance audit method further includes steps S301-S302: Step S301, sending the preliminary review result to the user, so that the user can verify according to the preliminary review result and return the verification result; First, the preliminary review results generated by the natural language processing model and the computer vision model are notified to the user via email or in-app messages. The user is provided with a verification option, allowing the user to view the marked content and give an option to indicate whether they agree with the preliminary review results. The user verifies these preliminary review results and transmits feedback and verification results to the system by selecting whether they agree or not.

[0058] The front-end page also provides a feedback form or button that users can click to report new types of violations. Users fill in a detailed description of the violation content, including but not limited to the violation type, screenshots or links to the violation content, and the specific location of the violation content.

[0059] Step S302, obtaining the verification result returned by the user, and sending the verification result to a pre-trained natural language processing model and a pre-trained computer vision model.

[0060] The system collects user feedback and appeal results on the preliminary review results, and confirms the addition of new illegal content or non-illegal content to the corresponding training data set based on user feedback for update training of natural language processing models and pre-trained computer vision models.

[0061] Through the above-mentioned embodiment method, user verification can not only help correct errors in the system, but also promote the system's rapid response and adaptability to new violations, which helps to continuously improve the accuracy and comprehensiveness of the audit system.

[0062] For example, to help understand the implementation process of the data compliance audit method obtained by combining this embodiment with the above-mentioned embodiment 1, please refer to Figure 3 , Figure 3 A brief flowchart of a data compliance audit method is provided, specifically: First, obtain enterprise WeChat data from enterprise WeChat, including but not limited to conversation data, circle of friends data, and customer note data. Then, classify these enterprise WeChat data to obtain text data, document files, audio files, and media files, and extract pure audio data and pure video data from the media files, where the media files include pictures and videos. Next, send the text data, document files, audio files, and pure audio data to a pre-trained natural language processing model for analysis to identify potential violations, such as sensitive words, illegal speech, etc. Send the media files and pure audio data to a pre-trained computer vision model to identify potential violations, such as pornography, violence, trademark infringement, etc.

[0063] The outputs of the two models are summarized to generate preliminary audit results, which are then sent to the account manager for professional review, and the account manager's feedback and complaint results are collected.

[0064] Subsequently, based on the feedback and appeal results from the account manager, new violations or appeals are confirmed and added to the corpus to update the training sets of the natural language processing model and the computer vision model. The updated training sets are used to iteratively optimize the natural language processing model and the computer vision model to improve the accuracy of the review.

[0065] Through the above-mentioned embodiment method, by combining the natural language model and the computer vision model, the enterprise WeChat content review task can be automated, reducing the need for manual intervention, thereby improving processing speed and efficiency. In addition to comprehensively covering the enterprise WeChat conversation content review function, this solution is further extended to the review field of enterprise WeChat Moments and customer notes, ensuring that all interactions and information records meet strict compliance standards. In this way, enterprises can not only maintain transparency and professionalism in daily communication, but also meet regulatory requirements in the management of shared content in Moments and personalized notes for customers, thereby protecting the corporate reputation and user data security in all aspects. At the same time, the model is iteratively optimized based on the feedback and complaint results of the account manager to further improve the accuracy of the review. In addition, this solution can also be widely used in data compliance audits in other fields and has broad application prospects.

[0066] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the data compliance review method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0067] This application also provides a data compliance audit device, please refer to Figure 4 , the data compliance audit device comprises: The classification module 10 is used to obtain enterprise WeChat data and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; The review module 20 is used to perform data compliance review on the different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results; The optimization module 30 is used to iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review results.

[0068] The data compliance audit device provided by this application adopts the data compliance audit method in the above embodiment, which can solve the technical problem of how to efficiently and accurately conduct compliance audits on various types of data in enterprise WeChat. Compared with the prior art, the beneficial effects of the data compliance audit device provided by this application are the same as the beneficial effects of the data compliance audit method provided by the above embodiment, and the other technical features in the data compliance audit device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0069] The present application provides a data compliance audit device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the data compliance audit method in the above-mentioned embodiment one.

[0070] Reference below Figure 5 , which shows a schematic diagram of the structure of a data compliance audit device suitable for implementing an embodiment of the present application. The data compliance audit device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The data compliance audit device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0071] like Figure 5 As shown, the data compliance audit device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the data compliance audit device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the data compliance audit device to communicate with other devices wirelessly or by wire to exchange data. Although the data compliance audit device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0072] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0073] The data compliance audit device provided by this application adopts the data compliance audit method in the above embodiment, which can solve the technical problem of how to efficiently and accurately conduct compliance audits on various types of data in enterprise WeChat. Compared with the prior art, the beneficial effects of the data compliance audit device provided by this application are the same as the beneficial effects of the data compliance audit method provided by the above embodiment, and the other technical features in the data compliance audit device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0074] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0075] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0076] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the data compliance audit method in the above-mentioned embodiment.

[0077] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0078] The computer-readable storage medium may be included in the data compliance audit device; or may exist independently without being assembled into the data compliance audit device.

[0079] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the data compliance audit device, the data compliance audit device: Acquire enterprise WeChat data, and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; Perform data compliance review on the different types of enterprise WeChat data by using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results; Based on the preliminary review results, the pre-trained natural language processing model and the pre-trained computer vision model are iteratively optimized.

[0080] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0081] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0082] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0083] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned data compliance review method, and can solve the technical problem of how to efficiently and accurately conduct compliance review of various types of data in enterprise WeChat. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the data compliance review method provided in the above-mentioned embodiment, and will not be repeated here.

[0084] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned data compliance audit method when executed by a processor.

[0085] The computer program product provided by this application can solve the technical problem of how to efficiently and accurately conduct compliance audits on various types of data in WeChat for Enterprises. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the data compliance audit method provided by the above embodiment, and will not be repeated here.

[0086] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A data compliance audit method, characterized in that: The method comprises: Acquire enterprise WeChat data, and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; Perform data compliance review on the different types of enterprise WeChat data by using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results; Based on the preliminary review results, the pre-trained natural language processing model and the pre-trained computer vision model are iteratively optimized.

2. The method according to claim 1, characterized in that The step of performing data compliance review on the different types of enterprise WeChat data by using a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary review results includes: For different types of enterprise WeChat data, select the corresponding pre-trained natural language processing model and / or pre-trained computer vision model to conduct data compliance review and obtain preliminary review results.

3. The method according to claim 2, characterized in that The different types of enterprise WeChat data include text data, pictures and video data. For the different types of enterprise WeChat data, the steps of selecting a corresponding pre-trained natural language processing model and / or a pre-trained computer vision model to perform data compliance review and obtaining a preliminary review result include: Processing text data using a pre-trained natural language processing model to extract illegal text content; and / or Process image and video data using pre-trained computer vision models to identify illegal visual content; The illegal text content and the illegal visual content are summarized to generate a preliminary review result.

4. The method according to claim 3, characterized in that Before the step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review result, the method further includes: Send the preliminary review result to the user, so that the user can verify it according to the preliminary review result and return the verification result; The verification result returned by the user is obtained, and the verification result is sent to a pre-trained natural language processing model and a pre-trained computer vision model.

5. The method according to claim 4, characterized in that The step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review result comprises: Based on the verification results returned by the users, the pre-trained natural language processing model and the pre-trained computer vision model are iteratively optimized.

6. The method according to claim 5, characterized in that The step of iteratively optimizing the pre-trained natural language processing model and the pre-trained computer vision model according to the verification result returned by the user includes: Confirm new illegal content or new appeal results based on the verification results returned by the user; In a pre-trained natural language processing model and a pre-trained computer vision model, iterative optimization is performed based on the new illegal content or the new complaint result.

7. A data compliance audit device, characterized in that: The data compliance audit device comprises: A classification module is used to obtain enterprise WeChat data and classify the enterprise WeChat data to obtain different types of enterprise WeChat data; An audit module, used to perform data compliance audit on the different types of enterprise WeChat data through a pre-trained natural language processing model and / or a pre-trained computer vision model to obtain preliminary audit results; An optimization module is used to iteratively optimize the pre-trained natural language processing model and the pre-trained computer vision model based on the preliminary review results.

8. A data compliance audit device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the data compliance audit method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the data compliance audit method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 of the data compliance audit are implemented.

Images