Data auditing method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a data auditing method and device, electronic equipment and storage medium, belonging to the technical field of artificial intelligence. The method comprises: obtaining a first auditing request sent by a request terminal; obtaining a terminal identifier of the request terminal and to-be-verified data according to the first auditing request; verifying the to-be-verified data according to a preset response condition to obtain a verification result; if the verification result is that the to-be-verified data meets the response condition, extracting a target auditing template from a preset auditing library according to the terminal identifier; wherein the target auditing template comprises data storage information and auditing content data; obtaining original data from the request terminal according to the data storage information; wherein the original data comprises at least one of the following data: original text data, original image data, and original video data; and auditing the original data according to the auditing content data to obtain a first auditing result. The embodiments of the present application can improve the content auditing efficiency and the auditing accuracy.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a data auditing method and apparatus, electronic device and storage medium. Background Technology

[0002] Currently, application content is reviewed manually. However, manual review is prone to inefficiency. Summary of the Invention

[0003] The main objective of this application is to provide a data review method, apparatus, electronic device, and storage medium, which aims to improve the efficiency of content review.

[0004] To achieve the above objectives, a first aspect of this application proposes a data auditing method, the method comprising:

[0005] Obtain the first review request sent by the requesting terminal;

[0006] Based on the first audit request, obtain the terminal identifier and the data to be verified of the requesting terminal;

[0007] The data to be verified is verified according to preset response conditions to obtain the verification result;

[0008] If the verification result indicates that the data to be verified meets the response conditions, then the target audit template is extracted from the preset audit database according to the terminal identifier; wherein, the target audit template includes data storage information and audit content data;

[0009] The raw data is obtained from the requesting terminal based on the data storage information; wherein the raw data includes at least one of the following: raw text data, raw image data, and raw video data;

[0010] The original data is reviewed based on the review content data to obtain the first review result.

[0011] In some embodiments, the data to be verified includes encrypted identity data, and the response condition includes a verification identifier;

[0012] The step of verifying the data to be verified according to preset response conditions to obtain a verification result includes:

[0013] The request identifier is obtained based on the encrypted identity data and the preset key;

[0014] The request identifier is verified based on the verification identifier to obtain the verification result.

[0015] In some embodiments, the data to be verified includes the number of concurrent requests for the original data, and the response conditions include a concurrency range;

[0016] The step of verifying the data to be verified according to preset response conditions to obtain a verification result includes:

[0017] Get the total number of concurrent connections;

[0018] The percentage data is obtained based on the number of concurrent requests and the total number of concurrent requests;

[0019] The percentage data is verified based on the concurrency range to obtain the verification result.

[0020] In some embodiments, the content to be reviewed includes a review algorithm. Before retrieving the target review template from a preset review database based on the terminal identifier if the verification result indicates that the data to be reviewed meets the response conditions, the data review method further includes constructing the target review template, specifically including:

[0021] Obtain the type of the original data;

[0022] The audit algorithm is extracted from a preset algorithm library based on the type of the original data;

[0023] Obtain the address parameters of the original data, and obtain the data storage information based on the address parameters;

[0024] The target review template is constructed based on the data storage information, the review algorithm, and the preset template identifier; wherein the template identifier matches the terminal identifier.

[0025] In some embodiments, the data auditing method further includes:

[0026] If the first review result is review timeout, review failure, or review abnormality, a prompt message is generated to prompt for manual review;

[0027] The prompt message is sent to the requesting terminal.

[0028] In some embodiments, the data auditing method further includes:

[0029] Obtain the second review request returned by the requesting terminal based on the prompt information;

[0030] Send the second review request to the reviewer and obtain the second review result returned by the reviewer based on the second review request;

[0031] The preset audit template is filled in based on the second audit result and the first audit result to obtain the target audit data.

[0032] In some embodiments, the data auditing method further includes:

[0033] Obtain the actual number of concurrent processes for the original data, and the processing time for the original data;

[0034] If the actual number of concurrent users is greater than the preset concurrency threshold and the processing time is less than the preset time threshold, then the audit result will be sent to the requesting terminal synchronously.

[0035] If the actual number of concurrent users is less than the concurrency threshold, or the processing time is greater than the processing time threshold, the audit result will be sent asynchronously to the requesting terminal.

[0036] To achieve the above objectives, a second aspect of this application provides a data verification device, the device comprising:

[0037] The request retrieval module is used to retrieve the first review request sent by the requesting terminal.

[0038] The first data acquisition module is used to acquire the terminal identifier and the data to be verified of the requesting terminal according to the first review request;

[0039] The response judgment module is used to verify the data to be verified according to preset response conditions and obtain the verification result.

[0040] An extraction module is used to extract a target audit template from a preset audit database if the verification result indicates that the data to be verified meets the response conditions; wherein the target audit template includes audit content data and data storage information;

[0041] The second data acquisition module is used to acquire raw data from the requesting terminal based on the data storage information; wherein the raw data includes at least one of the following: raw text data, raw image data, and raw video data;

[0042] The review module is used to review the original data based on the review content data to obtain the first review result.

[0043] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0044] To achieve the above objectives, a fourth aspect of the present application provides a storage medium, which is a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0045] The data auditing method, apparatus, electronic device, and storage medium proposed in this application verify the data to be verified. When the verification result shows that the data to be verified meets the response conditions, the target audit template is extracted from the preset auditing library through the terminal identifier. The original data is then processed according to the audit content data of the target auditing module, thereby realizing machine auditing of the original data. This avoids the problems of low auditing efficiency and low auditing accuracy caused by manual auditing in related technologies. Attached Figure Description

[0046] Figure 1 This is a flowchart of a data verification method provided in an embodiment of this application;

[0047] Figure 2 This is another flowchart of the data verification method provided in the embodiments of this application;

[0048] Figure 3 This is another flowchart of the data verification method provided in the embodiments of this application;

[0049] Figure 4 This is another flowchart of the data verification method provided in the embodiments of this application;

[0050] Figure 5 This is another flowchart of the data verification method provided in the embodiments of this application;

[0051] Figure 6 This is another flowchart of the data verification method provided in the embodiments of this application;

[0052] Figure 7 This is another flowchart of the data verification method provided in the embodiments of this application;

[0053] Figure 8 This is a schematic diagram of the data verification device provided in the embodiments of this application;

[0054] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0058] First, let's analyze some of the terms used in this application:

[0059] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0060] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). It is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0061] Content moderation is a detection technology based on images, text, audio, and video. It is used to detect whether images, text, audio, and video contain illegal content. By reviewing user-uploaded images, text, audio, and video, it determines the illegality of the uploaded content, thereby helping the content moderation request terminal reduce the risk of business violations. For image moderation, it can be applied to scenarios such as live video streaming, online stores, and website forums. In live video streaming scenarios, content moderation can monitor the live content of all rooms in real time, identify suspicious rooms, and issue warnings. In online store scenarios, content moderation can review images and videos uploaded by merchants and / or users, identify and issue warnings for non-compliant images and videos, and prevent the publication of illegal images and videos, thereby reducing the risk of business violations. For text moderation, it can be applied to e-commerce comment filtering, nickname verification for registration, media asset content verification, bullet screen comment verification, and real-time chat content verification. In addition, content moderation can also be applied to scenarios such as resolution detection. For clarity detection, content moderation can judge and quantify the clarity of enterprise forms, avoiding re-uploading and thus reducing labor costs; content moderation can also perform clarity detection on images uploaded to e-commerce review forums to avoid uploading blurry images, thereby ensuring the authenticity of uploaded images.

[0062] Concurrency: Also known as concurrency level, it refers to the number of requests a system can handle simultaneously, reflecting the system's load capacity. Concurrency can be obtained by analyzing the number of access logs on the requesting end within 1 second or through other methods. For example, when a webpage is viewed, the server establishes a link with the corresponding browser; each link represents one concurrent request. When a page contains multiple images, since the images are not displayed one by one, the server generates multiple links to send text and images simultaneously to improve browsing speed. Therefore, the more images on a page, the more concurrent connections the server can handle.

[0063] Microservices are a type of distributed architecture that advocates dividing a monolithic application into a set of small services that coordinate and cooperate with each other to provide end-user value. In a monolithic architecture, all services are integrated together. As business complexity increases, code coupling becomes increasingly high, hindering code upgrades and maintenance. Microservices break down a monolithic application into many independent projects based on business function modules, with each project completing a portion of the business functionality. For example, in an e-commerce system, there are order services, user functions, product services, and payment services. Implementing these modules using a monolithic architecture would increase coupling and development complexity. When developing with microservices, each service is treated as a monolithic application; that is, order services, user services, payment services, etc., are all microservices. The entire e-commerce system is composed of these microservices. Each microservice can be deployed in a cluster according to business needs, thereby reducing coupling between services and facilitating service maintenance and upgrades.

[0064] HTTP Request: HTTP, or Hypertext Transfer Protocol, is a response protocol used for communication between clients and servers. Common HTTP request methods include GET, POST, HEAD, PUT, DELETE, CONNECT, OPTIONS, and TRACE.

[0065] Currently, application content is reviewed manually. However, manual review is prone to problems such as low efficiency and low accuracy.

[0066] Based on this, embodiments of this application provide a data review method and apparatus, electronic device and storage medium, aiming to improve the efficiency and accuracy of content review.

[0067] The data auditing method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the data auditing method in this application is described.

[0068] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0069] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0070] The data auditing method provided in this application relates to the field of artificial intelligence technology. The data auditing method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the data auditing method, but is not limited to the above forms.

[0071] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0072] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0073] Figure 1 This is an optional flowchart of the data verification method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0074] Step S101: Obtain the first audit request sent by the requesting terminal;

[0075] Step S102: Obtain the terminal identifier and verification data of the requesting terminal according to the first verification request;

[0076] Step S103: Verify the data to be verified according to the preset response conditions to obtain the verification result;

[0077] Step S104: If the verification result is that the data to be verified meets the response conditions, then the target audit template is extracted from the preset audit database according to the terminal identifier; wherein, the target audit template includes data storage information and audit content data;

[0078] Step S105: Obtain raw data from the requesting terminal based on the data storage information; wherein, the raw data includes at least one of the following: raw text data, raw image data, and raw video data;

[0079] Step S106: Review the original data based on the review content data to obtain the first review result.

[0080] Steps S101 to S106 as shown in this embodiment of the application verify the data to be verified, and when the verification result shows that the data to be verified meets the response conditions, the target verification template is extracted from the preset verification library through the terminal identifier, and the original data is verified according to the verification content data of the target verification module, thereby realizing machine verification of the original data, thus avoiding the problems of low verification efficiency and low verification accuracy caused by manual verification in related technologies.

[0081] It should be noted that in the embodiments of this application and the following embodiments, the data review method is used as an example for explanation. The data review platform can be set in the terminal or in the server, and this application embodiment does not specifically limit it. The data review platform interacts with the requesting terminal through API (Application Program Interface) interfaces and other means to realize content review of the original data of the requesting terminal.

[0082] In step S101 of some embodiments, the requesting terminal is an application that needs to perform content review. This application can be in a monolithic architecture or a microservice developed using a distributed architecture; this embodiment of the application does not specifically limit this. The data review platform obtains the first review request sent by the requesting terminal through API interfaces or other means.

[0083] In step S102 of some embodiments, the first review request carries a terminal identifier and data to be verified. The terminal identifier is used to identify the requesting terminal and can be an International Mobile Equipment Identity (IMEI), a Mobile Equipment Identifier (MEID), or a coded ID uniformly assigned by the data review platform, etc. This application embodiment does not specifically limit this. The data to be verified is used to determine whether the first review request meets the corresponding conditions. It can be adaptively set according to content review requirements, and this application embodiment does not specifically limit this.

[0084] In step S103 of some embodiments, corresponding response conditions are pre-set according to the content review requirements. For example, when the content review requirement is that the requesting terminal has the authority to review the request, the response condition is a condition used to determine the request authority. The data to be reviewed is verified according to this response condition, and a corresponding verification result is obtained. This verification result indicates whether content review is performed on the original data of the requesting terminal. The original data refers to the objects in the requesting terminal that require content review, including original text data, original image data, original video data, etc.

[0085] Please see Figure 2 In some embodiments, the data to be verified includes encrypted identity data, the response condition includes a verification identifier, and step S102 may include, but is not limited to, steps S201 to S202:

[0086] Step S201: Obtain the request identifier based on the identity encryption data and the preset key;

[0087] Step S202: Verify the request identifier based on the verification identifier to obtain the verification result.

[0088] In step S201 of some embodiments, the response condition includes a verification identifier, which is used to determine whether the requesting terminal has the authority to call the data review platform for content review, thereby determining whether the first review request is a forged request. It is understood that when it is determined that the requesting terminal does not have the authority to call the data review platform for content review, the first review request sent by the requesting terminal is considered a forged request; when it is determined that the requesting terminal has the authority to call the data review platform for content review, the first review request sent by the requesting terminal is considered a genuine request. Therefore, the data to be verified includes encrypted identity data, which is data obtained by the requesting terminal through encryption calculation based on the verification identifier and verification key pre-allocated by the data review platform. The data review platform has a pre-set decryption key corresponding to the verification key. The data review platform decrypts the encrypted identity data using the decryption key to obtain the request identifier. It is understood that the request identifier has the same form as the verification identifier; that is, when the verification identifier is in encoded form, the request identifier is also in encoded form.

[0089] In step S202 of some embodiments, the data review platform has an identifier library that stores multiple verification identifiers. The data review platform searches and matches the identifier library according to the request identifier. When a verification identifier corresponding to the request identifier is found in the identifier library, it indicates that the verification identifier in the requesting terminal is an identifier assigned by the data review platform, thereby determining that the data to be verified meets the response conditions, that is, the requesting terminal has the authority to call the data review platform for content review. At this time, the data review platform responds to the review request and performs content review. When a verification identifier corresponding to the request identifier cannot be found in the identifier library, it indicates that the verification identifier in the requesting terminal is a forged identifier, thereby determining that the data to be verified does not meet the response conditions, that is, the requesting terminal does not have the authority to call the data review platform for content review. At this time, the data review platform may return a judgment result according to a preset strategy and / or generate a warning prompt, etc., which is not specifically limited in this embodiment of the application.

[0090] This application embodiment uses encrypted identity data to determine the authenticity of the first review request, thereby preventing the data review platform from reviewing content based on forged requests and thus preventing the data review platform's review resources from being arbitrarily accessed.

[0091] Please see Figure 3 In other embodiments, the data to be verified includes the number of concurrent requests for the original data, the response conditions include the concurrency range, and step S102 may include, but is not limited to, steps S301 to S303:

[0092] Step S301: Obtain the total number of concurrent connections;

[0093] Step S302: Obtain the percentage data based on the number of concurrent requests and the total number of concurrent requests;

[0094] Step S303: Verify the percentage data based on the concurrency range to obtain the verification results.

[0095] In step S301 of some embodiments, the data auditing platform obtains its total concurrency. It is understood that this total concurrency is used to characterize the maximum concurrency that the data auditing platform can handle. The total concurrency is affected by factors such as the number of CPU cores, memory capacity, network bandwidth, and hardware read / write speed of the data auditing platform. Therefore, the total concurrency can be obtained by pre-evaluating these influencing factors, or it can be obtained by evaluating these factors after receiving audit requests. This application embodiment does not specifically limit this.

[0096] In step S302 of some embodiments, the data to be verified includes the request concurrency count, which is obtained by the requesting terminal in advance by assessing its own content review needs. For example, when the requesting terminal is an application corresponding to an online store and the raw data includes raw image data, the request concurrency count can be obtained by analyzing the number of images uploaded by users within a certain period of time. It is understood that the request concurrency count can be the average concurrency count of the requesting terminal or the maximum concurrency count of the requesting terminal. However, in order to ensure that the data review platform can effectively allocate content review resources, in this embodiment of the application, the maximum concurrency count of the requesting terminal is used as the request concurrency count. The percentage of the request concurrency count to the total concurrency count is calculated based on the request concurrency count.

[0097] In step S303 of some embodiments, the data review platform can simultaneously review content from multiple different requesting terminals. Therefore, in order to respond to the first review requests from more requesting terminals, the content review resources for each requesting terminal need to be reasonably allocated, i.e., the concurrency of the reviewing terminal sending raw data needs to be limited. Specifically, a concurrency range is preset, and the concurrency range is compared with the percentage data. When the percentage data is not within the concurrency range, it indicates that the requesting terminal will occupy more content review resources, i.e., the data to be verified does not meet the response conditions. At this time, the data review platform will reject the first review request from the requesting terminal and return a rejection reason, etc. When the percentage data is within the concurrency range, it indicates that the request concurrency of the requesting terminal is within the load capacity of the data review platform, i.e., the data to be verified meets the response conditions. It can be understood that, in order to ensure that content review can be performed on each requesting terminal, when the percentage data is not within the concurrency range, the data review platform can delay the review of the requesting terminal according to a preset strategy, i.e., wait for other requesting terminals to complete their content reviews before responding to the first review request from the requesting terminal. It is understood that the concurrency range can be adaptively set according to the performance of the data audit platform, the number of request terminals with assigned permissions on the data audit platform, etc., and this application embodiment does not make specific limitations.

[0098] This application embodiment obtains the percentage data by combining the number of concurrent requests and the total number of concurrent requests, and obtains the verification result based on the preset concurrency range and the percentage data. This achieves the limitation on request terminals with a large number of concurrent requests, thereby ensuring that the data audit platform can respond to the first audit requests of more request terminals and improve the application scope of the data audit platform.

[0099] In step S104 of some embodiments, the data auditing platform has a preset auditing library, which includes multiple target auditing templates. Each target auditing template includes multiple data storage information, multiple auditing content data, and a template identifier for that target auditing template. Each auditing content data is matched with one data storage information. When the verification result indicates that the data to be verified meets the response conditions, the data auditing platform searches and matches multiple template identifiers in the auditing library based on the terminal identifier to filter and obtain the corresponding target auditing template.

[0100] In step S105 of some embodiments, the data storage information is used to store address parameters of pre-acquired raw data. When a first review request is received from the requesting terminal, and the data to be verified meets preset response conditions, the data review platform can read the corresponding raw data from the requesting terminal according to the address parameters stored in the data storage information. It is understood that the raw data refers to the object in the requesting terminal that needs content review. Raw data includes raw text data, raw image data, raw video data, raw audio data, etc., and this embodiment does not specifically limit its scope.

[0101] In step S106 of some embodiments, the target review template includes review content data, which includes algorithms required for content review of the original data. The original data is reviewed according to the algorithms in the review content data, and a first review result is generated based on the output of the corresponding algorithm and / or the processing result, thereby achieving machine review of the original data. It is understood that the review content data is pre-set according to the type of the original data; that is, when the original data is original text data, the review content data includes processing algorithms for preprocessing the original text data and recognition algorithms for recognizing the original text data. It is also understood that the number of review content data in the target review template is pre-set according to the number of original data for which the requesting terminal needs content review. For example, when the original data for which the requesting terminal needs content review includes original text data and original image data, the target review template corresponding to the requesting terminal includes two review content data sets, where the first review content data is used for content review of the original text data, and the second review content data is used for content review of the original image data. Depending on the actual review requirements, the original data can be reviewed sequentially according to the setting order of the first and second review content data; or, the original text data and the original image data can be reviewed simultaneously. This application embodiment does not specifically limit this.

[0102] Please see Figure 4 Before step S104 in some embodiments, the audit content data includes an audit algorithm. The data audit method provided in this application embodiment may also include, but is not limited to, the step of: constructing a target audit template, specifically including but not limited to steps S401 to S404:

[0103] Step S401: Obtain the type of the raw data;

[0104] Step S402: Extract the audit algorithm from the preset algorithm library according to the type of the original data;

[0105] Step S403: Obtain the address parameters of the original data, and obtain the data storage information based on the address parameters;

[0106] Step S404: Construct a target review template based on data storage information, review algorithm, and preset template identifier; wherein, the template identifier matches the terminal identifier.

[0107] In step S401 of some embodiments, the target review template is pre-set according to the content review requirements of the requesting terminal, and therefore the type of raw data is obtained according to these content review requirements. For example, when the requesting terminal is an application corresponding to an online store, its content review requirements include text review requirements, image review requirements, and video review requirements, that is, the raw data includes raw text data, raw image data, and raw video data. At this time, the type of raw data obtained includes text type, image type, and video type.

[0108] In step S402 of some embodiments, a corresponding review algorithm is extracted from a preset algorithm library according to the type of the original data. This review algorithm includes algorithms for content review of the original data, and may also include algorithms for preprocessing the original data. For example, when the type of the original data is an image, the extracted review algorithms include deep learning models capable of image recognition, and mean filtering algorithms, Gaussian filtering algorithms, etc., used for image preprocessing. It is understood that the preset algorithm library can be a database pre-loaded in the data review platform, or a database retrieved externally by the data review platform through API interfaces, etc. This application embodiment does not specifically limit this.

[0109] In step S403 of some embodiments, the storage address (i.e., address parameter) of the original data in the requesting terminal is obtained according to the content review requirements of the requesting terminal, and the data storage information is updated according to the address parameter. When the verification result indicates that the data to be verified meets the response conditions, the data review platform can obtain the original data according to the updated data storage information and update the input parameters of the review algorithm according to the original data, thereby realizing machine review of the original data according to the review algorithm.

[0110] In step S404 of some embodiments, a target review template is constructed based on multiple data storage information, multiple review algorithms, and a preset template identifier. Each data storage information corresponds to one review algorithm, and the template identifier is an identifier set according to the terminal identifier of the requesting terminal that requires content review, i.e., the template identifier matches the terminal identifier. For example, when the original data includes original text data, original image data, and original video data, the target review template includes three review algorithms and three data storage information. The first review algorithm is used for content review of text data; the second review algorithm is used for content review of original image data; and the third review algorithm is used for content review of original video data. Specifically, the data storage information including the address parameters of the original text data matches the first review algorithm; the data storage information including the address parameters of the original image data matches the second review algorithm; and the data storage information including the address parameters of the original video data matches the third review algorithm.

[0111] It is understandable that the input parameters of the review algorithm are updated based on the original data, meaning the original data is used as input data for the review algorithm, thereby enabling machine review of the original data according to the algorithm. It is also understandable that the first review result can be generated based on the output data of the review algorithm, or it can be a result generated during the algorithm's processing. For example, when the first review result is generated based on the output data of the review algorithm, the review algorithm could be an algorithm used to identify whether the original data contains illegal data; in this case, the first review result would be either "contains illegal data" or "does not contain illegal data." When the first review result is a result generated during processing, the first review result could be "original data could not be obtained," or "review algorithm malfunction," etc.

[0112] It is understandable that in other embodiments, different type identifiers can be set according to actual application needs. For example, the data review platform and the requesting terminal may pre-agree that identifier 1 represents the original data type as text, identifier 2 represents the original data type as image, and identifier 3 represents the original data type as video. In this case, the target review platform includes multiple review content data. Each review content data is used to review the content of one original data. Each review content data includes review algorithms corresponding to different types. That is, a review content data simultaneously includes review algorithms corresponding to text type, review algorithms corresponding to image type, and review algorithms corresponding to video type. When the requesting terminal includes multiple original data, the data review platform assigns a review content data to each original data according to the review request, and extracts the corresponding review algorithm from the review content data through the type identifier corresponding to the original data, thereby realizing machine review of the original data.

[0113] This application's embodiments construct a target review algorithm by using data storage information, review algorithms, and template identifiers. This enables the reading of raw data from the requesting terminal based on data storage information, and the machine review of that raw data according to the review algorithm. Furthermore, by obtaining raw data through data storage information, the method avoids setting different types of review algorithms for a single raw data set, thereby achieving a lightweight design for the target review template.

[0114] Please see Figure 5 In some embodiments, the data verification method provided in this application also includes, but is not limited to, steps S501 to S502:

[0115] Step S501: If the first review result indicates any of the following: review timeout, review failure, review abnormality, then generate a prompt message to prompt manual review;

[0116] Step S502: Send the prompt message to the requesting terminal.

[0117] In step S501 of some embodiments, in order to effectively review the original data, when the first review result is any one of review timeout, review failure, or review exception, a prompt message is generated. This prompt message is used to indicate that the original data needs to be manually reviewed. Review timeout indicates that when the review algorithm in the review content data performs machine review on the original data, the review algorithm does not output any data within a predetermined time; review failure indicates that the original data cannot be obtained, or the review algorithm outputs empty data; review exception indicates that the data review platform is malfunctioning, the original data is partially missing, or the original data format is incorrect. It is understood that the above explanations of review timeout, review failure, and review exception, as well as the circumstances under which the prompt message is generated based on the first review result, are merely illustrative. The above content can be adaptively adjusted according to the actual content review strategy, and this application embodiment does not specifically limit this.

[0118] It is understandable that in some embodiments, different methods for generating prompt messages can be set according to different types of raw data. For example, for text-type raw data, prompt messages can be generated in cases where the raw text data is empty, the length of the raw text data is greater than a preset length, or the raw text data cannot be recognized, based on the characteristics of the text. For image-type raw data, prompt messages can be generated in cases where the clarity of the raw image data is less than a preset threshold, or the raw image data recognition fails, based on the characteristics of the image.

[0119] In step S502 of some embodiments, a prompt message is sent to the requesting terminal via an API interface or other means to prompt the requesting terminal user to manually review the original data.

[0120] This application embodiment generates prompt information by indicating the first review result, such as review timeout, review failure, or review exception, thereby achieving a combination of machine review and human review and ensuring the effectiveness of the review of the original data content.

[0121] Please see Figure 6 In some embodiments, the data verification method provided in this application also includes, but is not limited to, steps S601 to S603:

[0122] Step S601: Obtain the second audit request returned by the requesting terminal based on the prompt information;

[0123] Step S602: Send the second review request to the reviewer and obtain the second review result returned by the reviewer based on the second review request;

[0124] Step S603: Fill in the preset audit template according to the second audit result and the first audit result to obtain the target audit data.

[0125] In step S601 of some embodiments, after the prompt information is sent to the requesting terminal, a second review request indicating that manual review is required is obtained from the requesting terminal.

[0126] In step S602 of some embodiments, the data auditing platform forwards the second audit request to the auditing end to prompt the auditors at the auditing end to manually review the original data. The auditing end generates a second audit result that has been completed by the auditors or uploaded during the audit process, and the data auditing platform obtains the second audit result through API interfaces or other means.

[0127] In step S603 of some embodiments, the data auditing platform also presets an audit template, and fills the audit template with content according to the first audit result generated by the audit algorithm and the second audit result generated by manual audit, thereby obtaining an audit report (i.e., target audit data).

[0128] The data review method provided in this application combines machine review with human review by performing manual review based on the second review request returned by the requesting terminal, thereby improving the accuracy of reviewing the original data content. Furthermore, by filling a preset review template with the first and second review results, an automatic review report is generated, facilitating subsequent verification of the content review process by the user.

[0129] Please see Figure 7 In some embodiments, the data verification method provided in this application also includes, but is not limited to, steps S701 to S703:

[0130] Step S701: Obtain the actual number of concurrent processes for the original data, and the processing time for the original data;

[0131] Step S702: If the actual number of concurrent users is greater than the preset concurrency threshold and the processing time is less than the preset time threshold, then the audit result will be sent to the requesting terminal synchronously.

[0132] Step S703: If the actual number of concurrent users is less than the concurrency threshold, or the processing time is greater than the processing time threshold, the audit result will be sent asynchronously to the requesting terminal.

[0133] In step S701 of some embodiments, the actual concurrency of the raw data sent by the requesting terminal and the processing time for content review of the raw data are obtained. It is understood that the methods for calculating the actual concurrency of different review content data can be the same or different. For example, multiple review content data may obtain their actual concurrency by calculating the number of connections to the raw data. Alternatively, for review content data that reviews image-type raw data, the raw data is obtained through synchronous calls. Therefore, the actual concurrency of the review content data is calculated based on the HTTP request, and the completion of a single HTTP call indicates the request is complete. For review content data that reviews video-type raw data, the raw data is obtained through asynchronous calls. Since the HTTP call in an asynchronous request returns the receiving result of the data review platform, not the review result, the actual concurrency of the review content data is calculated based on the review algorithm. It is understood that the method of obtaining raw data through synchronous or asynchronous calls can be pre-set according to the review requirements of the requesting terminal or set in other ways; this embodiment of the application does not specifically limit this.

[0134] In step S702 of some embodiments, if the actual number of concurrent requests is greater than a preset concurrency threshold and the processing time is less than a preset time threshold, it indicates that the review of the content data on the corresponding original data has the characteristics of low time consumption and high concurrency. In this case, the review result is returned to the requesting terminal through synchronous calling. It is understood that the specific values ​​of the concurrency threshold and the time threshold can be adaptively set according to actual needs, and the embodiments of this application do not impose specific limitations.

[0135] In step S703 of some embodiments, if the actual number of concurrent users is less than the concurrency threshold, or the processing time is greater than the time threshold, it indicates that the review of the content data has the characteristics of high time consumption or weak concurrency capability in the review of the corresponding original data. In this case, the review result is called back to the requesting terminal through asynchronous call.

[0136] This application embodiment determines the callback audit result by using the actual concurrency of the original data and the processing time of the audit content data on the original data, thereby ensuring the timeliness and accuracy of the audit result callback.

[0137] Please see Figure 8 This application also provides a data auditing device that can implement the above-described data auditing method. The device includes:

[0138] The request acquisition module 810 is used to acquire the first audit request sent by the requesting terminal.

[0139] The first data acquisition module 820 is used to acquire the terminal identifier and the data to be verified of the requesting terminal according to the first verification request;

[0140] The response judgment module 830 is used to verify the data to be verified according to preset response conditions and obtain the verification result.

[0141] The extraction module 840 is used to extract the target audit template from the preset audit database if the verification result shows that the data to be verified meets the response conditions; wherein, the target audit template includes audit content data and data storage information;

[0142] The second data acquisition module 850 is used to acquire raw data from the requesting terminal based on data storage information; wherein the raw data includes at least one of the following: raw text data, raw image data, and raw video data;

[0143] The review module 860 is used to review the original data based on the review content data and obtain the first review result.

[0144] It is evident that the content of the above data auditing method embodiments is applicable to the embodiments of this data auditing device. The specific functions implemented by the embodiments of this data auditing device are the same as those of the above data auditing method embodiments, and the beneficial effects achieved are also the same as those achieved by the above data auditing method embodiments.

[0145] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned data verification method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0146] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0147] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0148] The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the data verification method of the embodiments of this application.

[0149] The 903 input / output interface is used to implement information input and output.

[0150] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0151] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0152] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0153] This application also provides a storage medium, which is a computer-readable storage medium, storing a computer program that, when executed by a processor, implements the above-described data auditing method.

[0154] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0155] The data review method, apparatus, electronic device, and storage medium provided in this application embodiment achieve machine review of raw data through a review algorithm in the reviewed content data, thereby avoiding the problems of low review efficiency and low review accuracy caused by manual review in related technologies. Based on preset response conditions, the data to be verified is judged, realizing permission verification for the requesting terminal to call the data review platform for content review, and realizing the reasonable allocation of review resources of the data review platform. By generating prompt information for manual review through review results indicating review timeout, review failure, review exception, etc., a combination of machine review and manual review is achieved, thus ensuring the effectiveness of raw data content review.

[0156] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0157] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0158] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0159] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0160] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0161] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0162] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0163] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0164] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0165] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0166] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A data auditing method, characterized in that, The data verification methods include: Obtain the first review request sent by the requesting terminal; Based on the first audit request, obtain the terminal identifier and the data to be verified of the requesting terminal; The data to be verified is verified according to preset response conditions to obtain the verification result; If the verification result indicates that the data to be verified meets the response conditions, then the target audit template is extracted from the preset audit database according to the terminal identifier; wherein, the target audit template includes data storage information and audit content data; The raw data is obtained from the requesting terminal based on the data storage information; wherein the raw data includes at least one of the following: raw text data, raw image data, and raw video data; The original data is reviewed based on the review content data to obtain a first review result; The data to be verified includes encrypted identity data, and the response conditions include a verification identifier; the step of verifying the data to be verified according to the preset response conditions to obtain a verification result includes: The request identifier is obtained based on the encrypted identity data and the preset key; The request identifier is verified based on the verification identifier to obtain the verification result; The data to be verified includes the number of concurrent requests, and the response conditions include a concurrency range; the verification of the data to be verified according to the preset response conditions to obtain the verification result includes: Get the total number of concurrent connections; The percentage data is obtained based on the number of concurrent requests and the total number of concurrent requests; The percentage data is verified based on the concurrency range to obtain the verification result.

2. The data verification method according to claim 1, characterized in that, The audit content data includes an audit algorithm. Before retrieving the target audit template from the preset audit database based on the terminal identifier if the verification result indicates that the data to be verified meets the response conditions, the data audit method further includes constructing the target audit template, specifically including: Obtain the type of the original data; The audit algorithm is extracted from a preset algorithm library based on the type of the original data; Obtain the address parameters of the original data, and obtain the data storage information based on the address parameters; The target review template is constructed based on the data storage information, the review algorithm, and the preset template identifier; wherein the template identifier matches the terminal identifier.

3. The data verification method according to any one of claims 1 to 2, characterized in that, The data auditing method also includes: If the first review result is review timeout, review failure, or review abnormality, a prompt message is generated to prompt for manual review; The prompt message is sent to the requesting terminal.

4. The data verification method according to claim 3, characterized in that, The data auditing method also includes: Obtain the second review request returned by the requesting terminal based on the prompt information; Send the second review request to the reviewer and obtain the second review result returned by the reviewer based on the second review request; The preset audit template is filled in based on the second audit result and the first audit result to obtain the target audit data.

5. The data verification method according to any one of claims 1 to 2, characterized in that, The data auditing method also includes: Obtain the actual number of concurrent processes for the original data, and the processing time for the original data; If the actual number of concurrent users is greater than the preset concurrency threshold and the processing time is less than the preset time threshold, then the audit result will be sent to the requesting terminal synchronously. If the actual number of concurrent users is less than the concurrency threshold, or the processing time is greater than the processing time threshold, the audit result will be sent asynchronously to the requesting terminal.

6. A data verification device, characterized in that, The data auditing device is used to implement the data auditing method according to any one of claims 1 to 5, and the data auditing device comprises: The request retrieval module is used to retrieve the first review request sent by the requesting terminal. The first data acquisition module is used to acquire the terminal identifier and the data to be verified of the requesting terminal according to the first review request; The response judgment module is used to verify the data to be verified according to preset response conditions and obtain the verification result. An extraction module is used to extract a target audit template from a preset audit database if the verification result indicates that the data to be verified meets the response conditions; wherein the target audit template includes data storage information and audit content data; The second data acquisition module is used to acquire raw data from the requesting terminal based on the data storage information; wherein the raw data includes at least one of the following: raw text data, raw image data, and raw video data; The review module is used to review the original data based on the review content data to obtain the first review result.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the data auditing method according to any one of claims 1 to 5.

8. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the data auditing method according to any one of claims 1 to 5.

Images