Content publishing method and device, electronic equipment, storage medium and program product

Abstract

The invention provides a content publishing method and device, electronic equipment, a storage medium and a program product, which can be applied to the technical fields of artificial intelligence, content generation, content generation based on artificial intelligence and the like. The content publishing method comprises the steps that in response to content publishing operation, content to be published is generated according to an object set corresponding to configuration information indicated by the content publishing operation, the object set is created according to a commodity set, and the configuration information indicates at least one target channel identifier and a target number threshold value; according to the target channel identifier, determining whether there is a target subject for publishing the to-be-published content, the target subject being an account under a channel corresponding to the target channel identifier, and the number of the target subject satisfying a target number threshold; and in response to determining that the target main body exists, publishing the to-be-published content to the target main body.

Description

Technical Field

[0001] This disclosure relates to the technical fields of artificial intelligence, content generation, and AI-based content generation, and more specifically, to a content publishing method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] In the internet advertising industry, businesses can use advertising platforms to place content in the news feeds of mobile applications to attract users. For example, the content can be creative advertising materials presented in the form of images, text, or videos, used to guide users to click and complete product purchases or redirect them to shopping platforms.

[0003] However, in the process of realizing the present invention, the inventors found that the related technologies have at least the following problems: how to effectively improve the efficiency of content creation and realize cross-platform and cross-account batch content publishing are urgent problems to be solved. Summary of the Invention

[0004] In view of the above, this disclosure provides a content publishing method, apparatus, electronic device, storage medium, and program product.

[0005] According to one aspect of this disclosure, a content publishing method is provided, comprising: in response to a content publishing operation, generating content to be published based on an object set corresponding to configuration information indicated by the content publishing operation, wherein the object set is created based on a product set, and the configuration information indicates at least one target channel identifier and a target number threshold; determining, based on the target channel identifier, whether there exists a target entity for publishing the content to be published, wherein the target entity is an account under a channel corresponding to the target channel identifier, and the number of the target entities satisfies the target number threshold; and, in response to determining that the target entity exists, publishing the content to be published to the target entity.

[0006] According to another aspect of this disclosure, a content publishing apparatus is provided, comprising: a content generation module, configured to generate content to be published based on an object set corresponding to configuration information indicated by the content publishing operation in response to a content publishing operation, wherein the object set is created based on a product set, and the configuration information indicates at least one target channel identifier and a target number threshold; a subject determination module, configured to determine whether there exists a target subject for publishing the content to be published based on the target channel identifier, wherein the target subject is an account under a channel corresponding to the target channel identifier, and the number of the target subjects satisfies the target number threshold; and a content publishing module, configured to publish the content to be published to the target subject in response to determining that the target subject exists.

[0007] According to another aspect of this disclosure, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more instructions, wherein, when executed by the one or more processors, the one or more processors cause the one or more processors to perform the method as described in this disclosure.

[0008] According to another aspect of this disclosure, a computer-readable storage medium is provided having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the methods described in this disclosure.

[0009] According to another aspect of this disclosure, a computer program product is provided, which includes computer-executable instructions that, when executed, are used to perform the methods described in this disclosure. Attached Figure Description

[0010] The above and other objects, features and advantages of this disclosure will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0011] Figure 1 This illustration schematically shows a system architecture to which a content publishing method can be applied according to embodiments of the present disclosure;

[0012] Figure 2 A flowchart illustrating a content publishing method according to an embodiment of this disclosure is shown schematically;

[0013] Figure 3 This illustration shows an example diagram of a product information display interface according to an embodiment of the present disclosure;

[0014] Figure 4A This illustration schematically shows an example diagram of an implementation process of a publish-subscribe service according to an embodiment of the present disclosure;

[0015] Figure 4B An example schematic diagram of a first configuration interface according to an embodiment of the present disclosure is shown;

[0016] Figure 5A The illustration shows an example diagram of the process for generating a collection of objects according to an embodiment of the present disclosure;

[0017] Figure 5B The illustration shows an example schematic diagram of the process of generating an image set according to an embodiment of the present disclosure;

[0018] Figure 6A An example schematic diagram of an object upload interface according to an embodiment of the present disclosure is shown;

[0019] Figure 6BThe illustration shows an example schematic diagram of the generation of an image set according to another embodiment of the present disclosure;

[0020] Figure 7 An example schematic diagram of a second configuration interface according to an embodiment of the present disclosure is shown;

[0021] Figure 8 The illustration shows an example diagram of the process for generating content to be published according to an embodiment of the present disclosure;

[0022] Figure 9 An example schematic diagram of a grouping process according to an embodiment of the present disclosure is shown;

[0023] Figure 10 This illustration schematically shows an example diagram of an automated content publishing process according to an embodiment of the present disclosure;

[0024] Figure 11 A block diagram of a content publishing apparatus according to embodiments of the present disclosure is schematically shown; and

[0025] Figure 12 A block diagram of an electronic device suitable for implementing a content publishing method according to an embodiment of the present disclosure is shown schematically. Detailed Implementation

[0026] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0027] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0028] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0029] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0030] Content creation methods can include manual creation, tool-based batch creation, and programmatic recommendation. None of these methods are particularly effective in improving content creation efficiency or enabling cross-platform, cross-account batch content publishing.

[0031] To this end, this disclosure provides a content publishing method, apparatus, electronic device, storage medium, and program product, which can be applied to the fields of artificial intelligence, content generation, and AI-based content generation. The content publishing method includes: in response to a content publishing operation, generating content to be published based on an object set corresponding to configuration information indicated by the content publishing operation, wherein the object set is created based on a product set, and the configuration information indicates at least one target channel identifier and a target number threshold; determining, based on the target channel identifier, whether there exists a target entity for publishing the content to be published, wherein the target entity is an account under a channel corresponding to the target channel identifier, and the number of target entities meets the target number threshold; and in response to determining that a target entity exists, publishing the content to be published to the target entity.

[0032] Figure 1 The illustration schematically depicts a system architecture to which a content publishing method can be applied according to embodiments of this disclosure. It should be noted that... Figure 1 The examples shown are merely examples of system architectures that can be applied to the embodiments of this disclosure, in order to help those skilled in the art understand the technical content of this disclosure, but do not mean that the embodiments of this disclosure cannot be used in other devices, systems, environments or scenarios.

[0033] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between different devices.

[0034] It should be noted that the content publishing method provided in this embodiment can generally be executed by the server 105. Accordingly, the content publishing device provided in this embodiment can generally be located in the server 105.

[0035] Alternatively, the content publishing method provided in this embodiment of the disclosure can also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103. Correspondingly, the content publishing device provided in this embodiment of the disclosure can also be disposed in the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0036] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0037] It should be noted that the sequence numbers of the operations in the following methods are for descriptive purposes only and should not be considered as indicating the execution order of the operations. Unless explicitly stated otherwise, the method does not need to be executed in the exact order shown.

[0038] Figure 2 A flowchart illustrating a content publishing method according to an embodiment of this disclosure is shown schematically.

[0039] like Figure 2 As shown, the content publishing method 200 includes operations S210 to S230.

[0040] In operation S210, in response to the content publishing operation, content to be published is generated based on the object set corresponding to the configuration information indicated by the content publishing operation, wherein the object set is created based on the product set, and the configuration information indicates at least one target channel identifier and a target number threshold.

[0041] In operation S220, based on the target channel identifier, it is determined whether there is a target entity for publishing the content to be published. The target entity is an account under the channel corresponding to the target channel identifier, and the number of target entities meets the target number threshold.

[0042] In operation S230, in response to the determination that a target subject exists, the content to be published is published to the target subject.

[0043] A content publishing operation refers to an instruction action that triggers at least one of the processes of content generation and publishing. For example, a content publishing operation can be triggered by at least one of the following methods: a user clicks the content generation button on the front-end interface, a user clicks the content publishing button on the front-end interface, content generation is automatically triggered in response to scheduled task A, content publishing is automatically triggered in response to scheduled task B, a content generation request is received from an external source via API, a content publishing request is received from an external source via API, etc., without further limitation.

[0044] Configuration information refers to the set of information that controls the generation and publication of content. The method of obtaining configuration information can correspond to the triggering method of the content publication operation. For example, if the content publication operation is triggered by clicking a button on the front-end interface, a configuration information input box or selection box can be displayed on the front-end interface after the user clicks the button, allowing the user to input configuration information. Alternatively, if the content publication operation is triggered in response to a scheduled task, candidate configuration information can be pre-set for different tasks. When the scheduled task is triggered, the corresponding candidate configuration information is determined as the configuration information, etc., without further limitation.

[0045] An object set refers to a collection of material resources used to generate content. For example, an object set may include product images, promotional videos, and marketing copy. A product set refers to a collection of products to be promoted. For example, a product set could be a list of best-selling products selected by a pre-defined product selection algorithm, where each product has its own SKUID, name, price, and other information. Content to be published refers to advertising materials that have already been generated.

[0046] The method for generating content to be published based on an object set can be configured according to actual business needs and is not limited here. For example, a content template can be predefined, and the corresponding candidate objects from the object set can be automatically populated into the predefined content template according to the configuration information to generate the content to be published. Alternatively, user preferences can be analyzed in real time, and the combination strategy of each candidate object in the object set can be dynamically adjusted. Another option is to generate multiple content variations simultaneously, and select the optimal combination of candidate objects to generate the content to be published through small-scale testing.

[0047] The target channel identifier is the identifier of the channel on which the content to be published will be placed. The target number threshold refers to the maximum number of pieces of content that can be created. After generating the content to be published, the existence of a target entity for publishing the content can be determined based on the target channel identifier and the target number threshold. A target entity refers to an account on a specific advertising platform that is qualified to publish the content.

[0048] In one embodiment, a subject mapping can be obtained, and multiple candidate subjects in the subject mapping can be filtered according to preset filtering rules determined based on configuration information. The target subject can then be determined based on the filtering results. The preset filtering rules refer to predefined judgment conditions used to filter suitable target subjects. The subject mapping represents the correspondence between each candidate subject and subject attributes; a candidate subject may be, for example, an advertising account.

[0049] For example, the preset filtering rules may include a first filtering rule for filtering the channels to which the candidate subject belongs, and a second filtering rule for filtering the number of coarse-grained data and fine-grained data. The first filtering rule can be used to detect whether the channel to which the candidate subject belongs matches the channel configured in the task, and the second filtering rule can be used to detect whether the number of coarse-grained data is less than a threshold, and whether the number of fine-grained data is less than a threshold.

[0050] The filtering result refers to the output after applying preset filtering rules to the candidate subjects. For example, the filtering result can be a boolean value, where true indicates that the subject is available and false indicates that the subject is unavailable. The target subject refers to the available subject that has passed the filtering. If the boolean value corresponding to each candidate subject is false, the content publishing process can be restarted. If the filtering result is true, the candidate subject can be identified as the target subject for use in the subsequent content creation process.

[0051] In the embodiments of this disclosure, by using preset filtering rules, available entities that meet the requirements are automatically selected from multiple candidate entities. This achieves intelligent allocation and load balancing of content publishing, prevents creation failures caused by exceeding the entity limit, ensures the smooth execution of cross-channel batch delivery, and thus helps improve resource utilization and stability in the content publishing process. The specific method of publishing the content to be published to the target entities can be configured according to actual business needs and is not limited here. For example, the optimal target entity can be intelligently selected based on the publishing effects of different advertising platforms. Alternatively, the publishing effect can be tested within a small number of accounts first, and then large-scale publishing can be carried out after optimization. Alternatively, the same content to be published can also be published simultaneously to multiple accounts under different advertising platforms.

[0052] In the embodiments of this disclosure, by decoupling the content generation and publishing processes, end-to-end automation of content generation and publishing is achieved, reducing manual operation costs and improving the efficiency of content generation and publishing. During this process, a configurable content generation mechanism ensures the standardization and quality of the generated content to be published, while intelligent selection of target subjects ensures the reasonable use of the content. By automatically delivering the content to target subjects, it facilitates batch content publishing across platforms and accounts, improving the effectiveness of content publishing.

[0053] According to embodiments of this disclosure, the product set can be created in the following manner: in response to a second timed task, multiple candidate product information obtained from the data mapping corresponding to the content publishing operation are evaluated according to a preset evaluation index to obtain the evaluation value of each of the multiple candidate product information; the product set is determined based on the first N candidate product information in the sorted multiple candidate product information, where N is a positive integer.

[0054] The second scheduled task refers to an automated task that is triggered periodically to initiate the product selection process. The trigger interval for the second scheduled task can be configured according to actual business needs and is not limited here. For example, it can be set to run every 7 days or daily using the Spark task scheduling framework to periodically update the product collection. The product collection refers to the pool of products selected through the automated product selection process and can be used to create subsequent object collections. For example, the product collection can be a list of products stored in a database table.

[0055] Preset evaluation metrics refer to pre-defined product evaluation standards that can be used to quantify product potential. The specific preset evaluation metrics selected can be configured according to actual business needs and are not limited here. For example, preset evaluation metrics may include at least one of the following: conversion rate, unique visitor value (UV value), or return on investment (ROI). Conversion rate can be determined based on the number of completed orders and user visits to the product detail page; unique visitor value can be determined based on gross merchandise volume (GMV) and user visits; and ROI can be determined based on annual profit and total investment.

[0056] The data mapping corresponding to content publishing operations refers to the offline big data tables in the e-commerce system related to advertising placement. These tables store raw data of candidate products, such as product identifiers, categories, and traffic logs. Candidate product information refers to the product data to be evaluated extracted from the data mapping, which may include, for example, product identifiers, product names, main product titles, main product images, secondary product images, brand names, and category names at various levels.

[0057] The specific data mapping involved can be configured according to actual business needs and is not limited here. For example, the data mapping may include at least one of the following: offline big data Hive table 1 of orders in the e-commerce system, offline big data Hive table 2 of UVs of product detail pages in the e-commerce system, and other offline big data Hive tables 3 of the e-commerce system.

[0058] Hive Table 1 provides information such as the daily order time, SKU ID, product name, brand, product category, and product price for each product. Hive Table 2 provides traffic log information for browsing SKUs, used to calculate data such as page views (PV), unique visitors (UV), and clicks, which are then used to calculate conversion rates. Hive Table 3 assists in providing advertising information for products, such as impressions and clicks on advertising platforms.

[0059] Evaluation refers to the process of scoring each candidate product by calculating preset metrics. For example, in a Spark task, an SQL query is executed to calculate the conversion rate of each product as the evaluation value. The evaluation value is the numerical result of a product under preset evaluation metrics, such as a conversion rate of 0.05 or a UV value of 10.5 for a certain product. It should be noted that the higher the evaluation value, the more the selected product matches the business's expected product offerings. In one example, only the conversion rate can be used to calculate the evaluation value; alternatively, conversion rate can be prioritized first, and the average commercial value of unique visitors can be prioritized second, etc., without limitation here.

[0060] After obtaining the evaluation values ​​of multiple candidate products, the candidate products can be sorted in descending order based on these evaluation values. For example, this can be achieved using the SQL statement "ORDERBYconversion_rateDESC". After obtaining the sorted candidate products, the top N products can be written to the storage medium using an overwrite update method to ensure that the product set always contains the optimal products.

[0061] For example, if the product collection currently contains product IDs [1,2,3], and the latest processed and filtered product ID is [2,3,4], then the final product collection will contain products [2,3,4], which is an overwrite update. The purpose of an overwrite update is to always ensure that the selected products are the top N products that meet business expectations, so as to ensure that after the product content is created, it can bring the merchant the maximum number of orders and traffic.

[0062] In one example, the storage medium can be a traditional database, a large data table, or middleware such as ClickHouse. The value of N can be configured according to actual business needs and is not limited here. For example, N=100 can retrieve the top 100 products with the highest conversion rates.

[0063] In the embodiments of this disclosure, the automated execution of a second timed task and preset evaluation indicators enable efficient and objective screening of a massive number of candidate products, ensuring the generation of the optimal product set. This reduces manual intervention and product selection time, improves the accuracy and efficiency of content creation, and ensures that the product selection results always meet business expectations, which helps to support the accuracy and maximize the effect of subsequent content release.

[0064] In an example of creating a product collection, a big data Spark task can be run periodically to extract data from an offline Hive table and filter products.

[0065] For example, Structured Query Language (SQL) statements can be used to query product data that meets certain criteria from an offline Hive table, filtering out products that match the campaign objectives from a massive amount of product data. The filtered products are then scored to measure their campaign value. The scored products are then sorted from highest to lowest score, and the top N products are selected. Basic product information for each of the N products is obtained and stored in a product set.

[0066] The data-driven approach provided in this embodiment enables the automated filtering of a set of products that meet advertising requirements. Furthermore, a product information display interface allows business personnel to view and manage the selected product information for each filtered product.

[0067] According to embodiments of this disclosure, the product collection includes multiple product information items; multiple product information items are displayed through a product information display interface, wherein the product information display interface displays a configuration control and an object upload control; in response to the configuration control being triggered, a first configuration interface or a second configuration interface is displayed; and in response to the object upload control being triggered, an object upload interface is displayed.

[0068] A product information display interface is a graphical user interface used to display a collection of products. For example, a product information display interface can show automatically filtered product information in a list format.

[0069] Configuration controls are interactive elements in the product information display interface used to trigger task configuration. Clicking a configuration control will display either a first configuration interface or a second configuration interface. In one embodiment, the first configuration interface may refer to the interface used to configure the task of generating Artificial Intelligence Generated Content (AIGC) materials, and the second configuration interface may refer to the interface used to configure the task of creating advertising materials. For example, when a user clicks a configuration control, a JavaScript event listener can be used. If the user clicks "New" or "Edit," the second configuration interface can be loaded and displayed; if the user needs to configure an AIGC task, they can be redirected to the first configuration interface.

[0070] An object upload control is an interactive element in a product information display interface used to trigger the object upload function. Clicking the object upload control displays the object upload interface. In one embodiment, the object upload interface can refer to an interface for uploading manually created materials. For example, when a user clicks the object upload control, the object upload interface can be invoked and displayed through front-end event handling. This object upload interface can provide a file selection dialog box and a StockKeeping Unit (SKU) information input area, etc.

[0071] In the embodiments of this disclosure, configuration and upload functions are integrated through a unified product information display interface, providing users with a centralized operation entry point. This enables a fully visualized operation from product query and task configuration to object upload, simplifying the user operation path and improving the usability and operational efficiency of the content creation and publishing process.

[0072] Figure 3 An example schematic diagram of a product information display interface according to an embodiment of the present disclosure is shown.

[0073] like Figure 3 As shown, in embodiment 300 of the product information display interface, it may include a top functional area, a product information display area, and a pagination control display area.

[0074] The top functional area can include a product name input box, a SKUID input box, a search button, an export button, an upload materials button, and a view automation task button. The product name input box allows for fuzzy searching by product name to quickly locate products. The SKUID input box allows entering multiple SKUIDs for batch searching of product information. The search button triggers a search for product information, displaying a list of products that meet the criteria. The export button exports the current search results to a file for convenient offline analysis by business personnel. The upload materials button redirects to the material upload interface, supporting the uploading of manually created materials for products in the product pool. The view automation task button redirects to the automation task management page, allowing users to view the configuration and status of automation tasks related to the product pool.

[0075] The product information display area can be used to display product information for each product. For example, product information includes SKUID, product name, creative title, primary category, secondary category, tertiary category, brand, main title, subtitle, product status, etc. The pagination control display area supports paginated display of product information, facilitating the browsing of large-scale data.

[0076] According to embodiments of this disclosure, the configuration information includes first sub-configuration information and second sub-configuration information; operation S210 may include: in response to an object generation configuration operation for a first configuration interface, generating an object set based on a product set and the first sub-configuration information corresponding to the object generation configuration operation; in response to a content generation configuration operation for a second configuration interface, generating content to be published based on the object set and the second sub-configuration information corresponding to the content generation configuration operation.

[0077] The first configuration interface refers to the graphical interface used by users to configure object generation tasks. Users can perform object generation configuration operations here to input the first sub-configuration information. Object generation configuration operations refer to the actions of users entering parameters and saving them on the first configuration interface. For example, clicking the "Save" button triggers the AIGC material generation process based on the product collection and the first sub-configuration information.

[0078] The first sub-configuration information refers to the configuration parameters related to the generation of objects (i.e., materials). For example, the first sub-configuration information may include at least one of the following: the number of videos produced (vertical / horizontal), the number of image materials, the number of creative texts, and the effective start and end time of the task.

[0079] After the user enters and saves the first sub-configuration information through the first configuration interface, the task can be sent to the AIGC system through the message queue (MQ). The AIGC system automatically generates candidate objects based on the product set and configuration parameters using an algorithm model, and sends the candidate objects back to the object set for storage.

[0080] The second configuration interface is the graphical interface used by users to configure content generation tasks. Users can perform content generation configuration operations here to input second sub-configuration information. Content generation configuration operations refer to the actions of users entering parameters and saving them on the second configuration interface. For example, clicking the "Save" button triggers the ad material creation process based on the object set and the second sub-configuration information.

[0081] The second sub-configuration information refers to the configuration parameters related to the generation of content (i.e., advertising materials). For example, the second sub-configuration information may include at least one of the following: task name, priority, advertising promotion date, bidding mode, bid, daily budget, media placement resources, number of materials automatically created (i.e., number of plans, number of ad groups, number of creatives), and list of materials accounts to be created.

[0082] After the user enters and saves the second sub-configuration information through the second configuration interface, the task configuration in the enabled state can be read through a scheduled task; based on the object set and configuration parameters, the plan name and unit name are generated, and creative copy and materials are randomly matched to form creative data; finally, the media advertising platform API is called to create the plan, unit and creative, and the counter table and material status are updated to complete the generation of the content to be published.

[0083] In the embodiments of this disclosure, product selection, material generation and advertising material creation are linked through phased configuration and automated processes, realizing the generation of batch advertising materials for cross-media and multi-account use, reducing manual intervention and operation time, and improving the production efficiency and consistency of objects and content.

[0084] According to embodiments of this disclosure, the product set includes multiple product information items, and the object set includes a sub-object set for each product information item. Generating the object set based on the product set and the first sub-configuration information corresponding to the object generation configuration operation may include: constructing a publish-subscribe service by using the first sub-configuration information of each of the multiple product information items as the publisher and the object generation component as the subscriber; and subscribing to each first sub-configuration information item based on the publish-subscribe service to generate a sub-object set corresponding to the product information item based on the first sub-configuration information item.

[0085] A sub-object set refers to the collection of all materials generated for a single product. For example, for a product with product ID 123, its sub-object set might include 3 vertical videos, 2 horizontal images, and 5 creative copy.

[0086] An object generation component refers to an AIGC system used for automated material production. This component can generate candidate objects based on product information and configuration parameters using an algorithmic model. A publish-subscribe service refers to an asynchronous communication mechanism based on message brokers.

[0087] When a user creates and saves an AIGC production task in the first configuration interface, each product in the product collection and its corresponding first sub-configuration information can be encapsulated into a message and sent to the specified topic through a message middleware, thus forming a publisher; at the same time, the object generation component acts as a subscriber, continuously listening to messages in the topic.

[0088] In this scenario, the object generation component acts as a subscriber, listening to and consuming messages from the topic in real time via a message queue. Each time it receives a message containing product information and the first sub-configuration information, it triggers a material generation process. For each received message, the object generation component parses out the product information and the first sub-configuration information, calls its internal algorithm model to generate candidate objects of the specified type, quantity, and specifications, and then sends these candidate objects back to the material pool component via API, storing them in the corresponding database table to form a collection of sub-objects for that product.

[0089] In the embodiments of this disclosure, the product configuration information and the object generation component are decoupled and processed asynchronously through a publish-subscribe service, enabling the object generation tasks of massive products to be executed in parallel and reliably, thereby improving throughput and processing efficiency. At the same time, the persistence mechanism of the message queue ensures the reliability of task execution and guarantees the stability and efficiency of the automated content generation process.

[0090] Figure 4A The illustration shows an example schematic diagram of an implementation process of a publish-subscribe service according to an embodiment of the present disclosure.

[0091] like Figure 4A As shown, in embodiment 400A of the implementation process of the publish-subscribe service, AIGC automated material production tasks can be created or modified through the user interface, and the task information can be sent to the AIGC system to generate creative materials.

[0092] For example, product information from the product pool can be displayed through a product information display interface. Users can click the "Create Task" or "Modify Task" button on the product information display interface to enter the task creation or modification page, triggering the task configuration process and allowing users to define task parameters. Users can fill in the relevant configuration information for the task on the task creation page to provide clear task requirements for the AIGC system. The entered task information is then saved to the database.

[0093] Task information is extracted from the database, packaged into a message body, and sent to a message middleware. The message middleware enables asynchronous transmission of task information, triggering the task execution process and notifying the AIGC system to begin material creation. The object generation component can subscribe to and consume task information from the message middleware. The task information is then stored in the object generation component's internal database or temporary cache.

[0094] Figure 4B An example schematic diagram of a first configuration interface according to an embodiment of the present disclosure is shown.

[0095] like Figure 4BAs shown, in the first configuration interface 400B, users can fill in task information to define the number of videos, images, and text to be generated for each product, as well as the effective time range of the task. After the task is submitted, the material generation task can be automatically executed according to the configuration.

[0096] For example, the first configuration interface 400B may include a task name input box, a video production count input box, an image production count input box, a text production count input box, and a task validity period input box.

[0097] The Task Name input box allows you to enter a unique identifier for each task, distinguishing different material production tasks and facilitating task management and retrieval. The Video Count input box allows you to enter the number and type of video materials to be generated for each product. The Image Count input box allows you to enter the number and type of image materials to be generated for each product. The Copy Count input box allows you to enter the number of creative copy texts to be generated for each product. The Task Validity period input box allows you to enter the start and end dates of the task, limiting the execution time and ensuring completion within the specified time. After configuration, you can click the Save button to save the entered task information and submit the task.

[0098] According to embodiments of this disclosure, a first configuration interface displays an object configuration control, which controls the object type and number of objects generated. The object type includes at least one of image type and text type, and the sub-object set includes at least one of image set and text set. Generating a sub-object set corresponding to product information based on the first sub-configuration information may include: generating a text object based on the number of objects for the text type and storing the text object in the text set; generating a first candidate object based on the number of objects for the image type; and storing the first candidate object as an image object in the image set in response to the successful execution of the object storage service for the candidate object.

[0099] The object configuration control refers to the input component in the first configuration interface used to set object generation parameters. Object type refers to the type of material generated by AIGC, which can include image and text types. Furthermore, image types can be divided into picture and video types. Object count refers to the number of objects to be generated for each object type. For example, the number of text objects is 5, and the number of image objects is 10.

[0100] For text-type items, the object generation component can generate a specified number of text objects based on product information using an algorithm model, according to the number of text-type objects set in the first sub-configuration information. These text objects are then stored in a text collection via API calls, with multiple text objects separated by newline characters (\\n). The text collection refers to the creative copy portion of the text-type items within the sub-object collection. For example, generating 5 creative copy titles for a specific product.

[0101] For image types, the object generation component can generate a first candidate object of a corresponding quantity and specifications based on the product information, according to the number of objects for the image type set in the first sub-configuration information and using an algorithm model.

[0102] Object storage service refers to the API interface service provided by the media pool component, used to receive and store AIGC-generated media. An image collection refers to the portion of a sub-object collection that includes image and video objects. For example, a collection consisting of 3 video files and 2 images generated for a product. After obtaining the first candidate object, it can be uploaded to the Object Storage Service (OSS) and an HTTP URL obtained, and then stored as an image object in the image collection.

[0103] In the embodiments of this disclosure, fine-grained control over the generation of AIGC materials is achieved through typed configuration controls. The number of materials to be generated is set independently for text and image types, and the corresponding generation and storage processes are executed. This enables parallel and standardized production of multiple types of materials, ensuring that the generated objects meet the requirements for content publishing. At the same time, the integrity and availability of the generated objects are ensured through the verification mechanism of the object storage service, providing a reliable data foundation for the large-scale automated creation of subsequent content.

[0104] Figure 5A The illustration shows an example schematic diagram of the process for generating a collection of objects according to an embodiment of the present disclosure.

[0105] like Figure 5A As shown, in Example 500A of generating object collections, product information can be obtained from the product collection through an automated component, and video, image and creative copywriting materials can be generated using AIGC technology. Finally, the generated materials are stored in the material pool table for use in the creation of subsequent advertising materials.

[0106] For example, a scheduled task can be used to trigger the generation of objects, ensuring that the material generation operation is automatically executed within a specified time. Task information is read from the task configuration, and product information is retrieved from the product collection via an API interface. For each product in the product collection, the material generation step is executed separately.

[0107] In the content generation stage, the AIGC model can be invoked to generate corresponding content based on product information and characteristics, such as video objects, image objects, and text objects. The generated video and image objects can be returned to an image collection via an API interface. In a specific example, the storage structure of the image collection is shown in Table 1 below.

[0108] Table 1

[0109]

[0110] The generated text object can be returned to the text collection via the API interface. In a specific example, the storage structure of the text collection is shown in Table 2 below.

[0111] Table 2

[0112]

[0113] After the materials for the current product are generated, you can cycle to the next product until the materials for all products are generated.

[0114] Figure 5B The illustration shows an example schematic diagram of the process for generating an image set according to an embodiment of the present disclosure.

[0115] like Figure 5B As shown, in embodiment 500B of generating image sets, video and image materials generated by the AIGC system can be received by an automated component and stored in a file storage system and database to ensure that the materials can be used in subsequent advertising material creation processes.

[0116] For example, video data and image information generated by the AIGC system can be received via API. The received video data and image information are parsed to structure the material data for easier subsequent storage and management. The parsed video files are then uploaded to an object storage service to ensure secure storage and efficient access. If the upload fails, the process ends; if the upload is successful, key information about the video materials is written to a data table to ensure that the metadata of the video materials is stored in the database for later use in creating advertising materials. Simultaneously, key information about the image materials is also written to a data table to ensure that the metadata of the image materials is stored in the database for later use in creating advertising materials.

[0117] According to embodiments of this disclosure, the content publishing method may further include the following operations: in response to a second candidate object uploaded via an object upload interface, verifying the object name of the second candidate object according to a preset verification rule to obtain a verification result, wherein the preset verification rule defines at least one of a format rule for the object name and a matching rule for the product identifier in the object name; in response to the verification result indicating that the object name passes the verification, storing the second candidate object as an image object in an image collection.

[0118] The object upload interface is a graphical interface for manually uploading objects. This interface provides an object selection area and an upload button. Secondary candidate objects refer to media files manually uploaded by the user through the object upload interface. Examples include a video file named "124343_a.mp4" and an image file named "124343_b.jpg" uploaded locally by the user.

[0119] Predefined validation rules refer to a set of predefined rules used to validate uploaded material filenames. These rules can include formatting rules and product identifier matching rules. The object name refers to the filename of the uploaded object. Formatting rules specify the naming conventions the filename must conform to. For example, a formatting rule could be SKUID_xxxx, which indicates that the uploaded object's filename includes two parts separated by underscores: the product identifier at the beginning and the custom description and extension at the end. The product identifier is the unique identifier embedded in the object name. Matching rules verify whether the product identifier actually exists. For example, a matching rule could be to check if the product identifier in the filename exists.

[0120] After a user uploads a second candidate object through the object upload interface, the backend receives the second candidate object and parses its object name. First, it checks if the object name conforms to the format rules. Then, it extracts the product identifier from the object name and can query the product set by calling the validation interface to determine if the product identifier matches the matching rules.

[0121] The validation result refers to the output after validating the object name. If the object name of the second object passes the format rule validation and the product identifier passes the matching rule validation, then the validation result indicating that the object name has passed the validation can be determined. If the object name of the second object fails the format rule validation and / or the product identifier fails the matching rule validation, then a prompt message can be entered for the user.

[0122] In one example, the validation interface could be a `Boolean checkSkuExist(String SKUID)` function, which retrieves the SKUID and returns a boolean value. For instance, returning `true` indicates that the object name passed validation, while returning `false` indicates that the object name failed validation.

[0123] When the verification result indicates that the object name passes the verification, the second candidate object can be uploaded to the object storage service, the HTTP URL access address can be obtained, and the second candidate object can be stored as an image object in the image collection.

[0124] In the embodiments of this disclosure, preset format rules and matching rules ensure the accurate association between manually uploaded objects and product sets, effectively preventing the upload of invalid or erroneous objects and ensuring the consistency and availability of object data. At the same time, the automated verification and storage process reduces the cost of manual review, provides a reliable data foundation for the automated creation of subsequent content, and improves the robustness and processing efficiency of the entire content generation and publishing process.

[0125] Figure 6A An example schematic diagram of an object upload interface according to an embodiment of the present disclosure is shown.

[0126] like Figure 6A As shown in the object upload interface 600A, users can manually upload the images or video materials required for advertising through this interface.

[0127] The 600A object upload interface displays a media type selection control, a media selection control, and a preview control. The media type selection control allows users to select either image or video media for upload. The media selection control opens a file selection dialog box to choose local media files. The preview control displays a preview image of the uploaded media, supporting previews of multiple images or video covers.

[0128] Figure 6B The illustration shows an example schematic diagram of the generation of an image set according to another embodiment of the present disclosure.

[0129] like Figure 6B As shown, in embodiment 600B of generating an image collection, a user can manually upload advertising materials through an object upload interface and store them in the image collection.

[0130] For example, users can log in through authentication to ensure that only authorized users can upload materials. Product information from the product pool is displayed for user reference. Users can click the "Upload Materials" button on the product information display interface to enter the object upload interface. In the object upload interface, users can select the type of material to upload and choose prepared local material files.

[0131] The process checks if the uploaded file's name conforms to the skuid_xxx specification to ensure the filename is compliant. If this fails, an error message "Does not conform to naming conventions" is displayed, and the process ends. If successful, it checks if the skuid in the filename exists in the product collection to ensure the material is associated with a product in the collection. If this fails, an error message "SKU does not exist" is displayed, and the process ends. If successful, the uploaded material is stored in the image collection and written to the database table.

[0132] According to embodiments of this disclosure, the second sub-configuration information includes priority information; generating content to be published based on the object set and the second sub-configuration information corresponding to the content generation configuration operation may include: in response to a first scheduled task, filtering the second sub-configuration information according to the execution status and priority information corresponding to the second sub-configuration information to obtain target sub-configuration information; and generating content to be published based on the object set and the target sub-configuration information.

[0133] The first scheduled task refers to a periodically triggered automated content creation process. The trigger interval for the first scheduled task can be configured according to actual business needs and is not limited here. For example, it can be set to run every 7 days or daily using the Spark task scheduling framework to trigger the content creation task periodically.

[0134] The second sub-configuration information refers to the set of configuration parameters related to the ad creative creation task. For example, the second sub-configuration information may include task name, priority information, ad promotion date, bidding mode, bid, daily budget, channel placement resources, number of automatically created content items, and list of main content items to be created.

[0135] Priority information refers to the priority level parameters used to determine the execution order of tasks, controlling the execution order of multiple tasks. For example, priority information can be data in the range of [1~10], with smaller numbers indicating higher priority. Execution status refers to the current state of an automated task. For example, execution status can include enabled or paused status.

[0136] The advertising promotion date refers to the start and end dates of the ad campaign when the ad is created. For example, if the time period is entered as 20250310~20250311, the ad will be effective from March 10th to March 11th and will participate in the bidding.

[0137] Bidding models may include at least one of the following: Cost Per Mille (CPM), Cost Per Click (CPC), Cost Per Action (CPA), Optimized Cost Per Click (oCPC), Optimized Cost Per Mille (oCPM), etc.

[0138] Bid is the price offered in the auction, expressed in yuan. Daily budget refers to the daily budget allocated to an advertising campaign or unit; exceeding this amount will automatically pause ad delivery. Channel placement resources can involve ad placements across multiple channels, used to determine the placement of created content to be published.

[0139] The number of automatically created content items can be categorized into three levels: plans, units, and creatives. The plan count represents the number of plans allowed per subject, the unit count represents the number of items allowed per plan, and the creative count represents the number of items allowed per unit. The content subject list refers to the target subjects for automated content creation.

[0140] When the first scheduled task is triggered, all second sub-configuration information can be read and filtered based on execution status and priority. The specific filtering method can be configured according to actual business needs and is not limited here. For example, tasks in the enabled state can be filtered first, then sorted by priority from high to low, and the task at the top of the list can be selected as the target sub-configuration information. The target sub-configuration information refers to the automated task configuration to be executed after filtering; that is, the configuration information of the enabled task with the highest priority selected from all tasks.

[0141] In the embodiments disclosed herein, an orderly automated processing in a multi-task environment is achieved by combining a timed triggering mechanism with priority information, ensuring that high-priority tasks are executed first; at the same time, invalid operations are avoided based on status filtering, resource allocation is optimized, the efficiency and reliability of content creation are improved, the timely delivery of key businesses is guaranteed, and intelligent management and large-scale operation of the content publishing process are realized.

[0142] Figure 7 An example schematic diagram of a second configuration interface according to an embodiment of the present disclosure is shown.

[0143] like Figure 7 As shown, in the second configuration interface 700, users can configure and submit tasks for automatically creating advertising materials.

[0144] The second configuration interface 700 displays the task name input box, priority input box, advertising promotion date input box, bidding mode input box, bid input box, daily budget input box, media placement resource selection box, automatic material creation quantity input box, and material account list input box.

[0145] The Task Name input box is used to enter a unique identifier for the task, distinguishing different automated tasks and facilitating task management and retrieval. The Priority input box is used to enter the task's priority, ensuring that higher-priority tasks are executed first. The Ad Promotion Date input box is used to enter the time range for ad campaign execution, limiting the ad campaign's duration and ensuring that the ad takes effect as planned.

[0146] The Bidding Mode input box is used to enter the bidding mode for ad placement, determining the billing method and delivery strategy. The Bid input box is used to enter the single bid amount for the ad, defining the bidding strategy. The Daily Budget input box is used to enter the daily budget limit for the ad, controlling daily spending. The Media Placement Selection box is used to select the media and placement resources for ad placement. The Automatic Material Creation Quantity input box is used to enter the number of campaigns, ad groups, and creatives, controlling the number of materials automatically created. The Material Creation Account List input box is used to enter the account names for which ad materials will be created, specifying the target accounts for ad material creation.

[0147] According to embodiments of this disclosure, the target sub-configuration information includes a product identifier, and the object set includes an image set and a text set. Generating content to be published based on the object set and the target sub-configuration information may include: obtaining multiple candidate images from the image set and multiple candidate texts from the text set based on the product identifier; and generating content to be published based on the multiple candidate images and multiple candidate texts.

[0148] A product identifier refers to a unique identification code for a product. After obtaining the target sub-configuration information, based on the product identifiers involved in the target sub-configuration information, firstly, candidate images that meet the conditions are queried from the image set. These candidate images are the image types that meet the creation conditions found from the image set. Secondly, candidate texts corresponding to the product identifiers are queried from the text set. These candidate texts are the text types that meet the creation conditions found from the text set.

[0149] The query conditions for image types and text types can be configured according to actual business needs and are not limited here. For example, the query condition for image types can be is_delete=0 AND use_flag=0, and the top 1000 candidate images sorted in descending order of creation time; the query condition for text types can be SKUID=product identifier AND is_delete=0, that is, querying data that has not been deleted.

[0150] After obtaining multiple candidate images and multiple candidate texts, the candidate images and candidate texts can be combined to generate content to be published. The specific combination method can be configured according to actual business needs and is not limited here. For example, the candidate images can first be grouped according to the product identifier and the number of creatives set in the automation task; then, one candidate text can be randomly selected for each image group; next, coarse-grained identifiers and fine-grained identifiers can be generated; finally, a multi-level tree structure of content to be published can be formed based on these elements.

[0151] In the embodiments of this disclosure, matching candidate images and candidate texts are automatically obtained through product identifiers, and the candidate images and candidate texts are intelligently combined, thereby realizing the automated construction of content to be published, ensuring the relevance and diversity of images and texts, reducing the workload of manual matching and combination, and improving the efficiency and quality of content generation.

[0152] Figure 8 The illustration shows an example diagram of the process for generating content to be published according to an embodiment of this disclosure.

[0153] like Figure 8 As shown, in embodiment 800 of generating content to be published, advertising materials can be created through automated components and published to a designated advertising account.

[0154] For example, an automated material creation process can be triggered through a scheduled task mechanism to ensure that tasks are executed automatically at predetermined times. The second-level configuration information of the automated tasks is read. Tasks to be processed are filtered based on their execution status to ensure that only tasks in an executable state are processed. Tasks are sorted and filtered according to their priority to obtain target sub-configuration information, ensuring that high-priority tasks are executed first.

[0155] Read multiple candidate images from the image set. Group the candidate images into multiple image groups. Generate coarse-grained and fine-grained labels for each image group. Read available candidate text from the text set. Randomly combine candidate images and candidate text to generate diverse ad creative content, thus generating the content to be published.

[0156] The system iterates through the subject map, filtering each candidate subject based on the target number threshold and the number of candidates for each subject in the subject map to determine if it is an available subject. If it is not an available subject, it is skipped, and the next candidate subject is processed; if it is an available subject, it is designated as the target subject. The content to be published is then published to the target subject by invoking the content publishing component for that subject.

[0157] In addition, after the content to be published is completed, the number of candidate materials used and the usage status of the materials can be updated. After completing the creation of the current batch of materials, the next batch of objects can be processed.

[0158] According to embodiments of this disclosure, the target sub-configuration information further includes an object number threshold; the content publishing method may further include: grouping multiple candidate images according to the object number threshold to obtain multiple image groups, wherein the number of candidate images in each image group is less than or equal to the object number threshold; and generating a content identifier for each image group based on the product identifier.

[0159] The object number threshold refers to the maximum number of creative ideas allowed to be created in each unit. An image group is a set formed by grouping candidate images according to the object number threshold. For example, if there are 10 candidate objects corresponding to the same product label and the object number threshold is 3, after grouping, there will be 4 image groups, of which 3 image groups each contain 3 candidate images, and the remaining image group contains 1 candidate image.

[0160] A content identifier is a unique identifier name generated for content. For each image group, a corresponding content identifier can be generated based on the product identifier and preset naming rules. In one embodiment, the content identifier can include coarse-grained identifiers and fine-grained identifiers, with the product identifier corresponding to multiple levels of category names. Coarse-grained identifiers refer to the naming identifiers at the planning level. Fine-grained identifiers refer to the naming identifiers at the unit level. Category names refer to the classification information of products in the e-commerce system, and can include multiple levels, such as first-level categories, second-level categories, and third-level categories.

[0161] In the embodiments of this disclosure, candidate images are intelligently grouped according to a threshold number of objects to ensure that the number of creative ideas in each unit meets the requirements. At the same time, a unique content identifier is generated for each image group through standardized naming rules, which realizes the standardization and batch management of content generation, effectively prevents creation failure caused by exceeding the unit limit, and improves the stability and management efficiency of content delivery.

[0162] According to embodiments of this disclosure, generating a content identifier based on a product identifier may include: generating a coarse-grained identifier based on the product identifier and the category name at the preceding level; and generating a fine-grained identifier based on the product identifier and the category name at the following level.

[0163] Preset naming rules can include coarse-grained naming rules for the coarse-grained level and fine-grained naming rules for the fine-grained level. Specific coarse-grained and fine-grained naming rules can be configured according to actual business needs and are not limited here.

[0164] For example, coarse-grained naming rules can be determined based on the category names at the previous level, which refer to the category names of higher-level products. In a specific example, taking the category names of higher-level products as including first-level and second-level categories, the coarse-grained naming rule can be SKUID-first-level category name-second-level category name-date-4-digit custom English label-serial number, where the meaning of each component is shown in Table 3 below.

[0165] Table 3

[0166]

[0167] During the generation of coarse-grained identifiers, the product basic information query interface can be called through the product identifier to obtain the product's first-level category name and second-level category name. Then, according to the coarse-grained naming rules mentioned above, this information is combined with the date, custom English label, and serial number to generate the coarse-grained identifier. For example, 123-Digital-Computer-250721-AUTO-1.

[0168] For example, fine-grained naming rules can be determined based on the category name at the next lower level, which refers to the category name of the lower-level product. In a specific example, taking the lower-level product category name as including a third-level category, the fine-grained naming rule can be SKUID-third-level category name-bidding mode-creation time-4-digit custom English label-serial number, where the meaning of each component is shown in Table 4 below.

[0169] Table 4

[0170]

[0171] During the generation of fine-grained identifiers, the third-level category name can be obtained from the same product identifier. Combined with information such as the bidding mode and creation time in the automated task configuration, fine-grained identifiers are generated according to the fine-grained naming rules. For example, 123-Hard Disk-OCPM-250721171205-AUTO-1.

[0172] In the embodiments of this disclosure, a hierarchical identifier generation mechanism is used to achieve clear distinction and effective management between coarse-grained and fine-grained levels, enhancing the identifiability and traceability of the content to be published. At the same time, standardized naming rules ensure the uniqueness and standardization of identifiers when creating in batches, thereby improving the management efficiency of content publishing.

[0173] Figure 9 An example schematic diagram of a grouping process according to an embodiment of the present disclosure is shown.

[0174] like Figure 9 As shown, in embodiment 900 of the grouping process, taking video materials including 123_1.mp4, 123_2.mp4, 123_3.mp4 and 123_4.mp4, 456_1.mp4 and 456_2.mp4, and image materials including 123_pic1.jpg, 123_pic2.jpg, 456_pic1.jpg and 789_pic1.jpg, with an object number threshold of 3 as an example, the process of grouping and managing materials according to the product identifier is explained.

[0175] For example, you can first filter the materials according to the product identifier, that is, filter the following files corresponding to product identifier 123: 123_1.mp4, 123_2.mp4, 123_3.mp4, 123_4.mp4, 123_pic1.jpg, and 123_pic2.jpg; the following files corresponding to product identifier 456: 456_1.mp4, 456_2.mp4, and 456_pic1.jpg; and the following file corresponding to product identifier 789: 789_pic1.jpg.

[0176] Since the amount of material corresponding to product identifier 123 exceeds the threshold for the number of objects, these materials can be further divided. For example, image group 910 corresponding to product identifier 123 can be generated based on 123_1.mp4, 123_2.mp4, and 123_pic1.jpg, and image group 920 corresponding to product identifier 123 can be generated based on 123_3.mp4, 123_4.mp4, and 123_pic2.jpg.

[0177] Furthermore, since the material corresponding to product identifier 456 does not exceed the object number threshold, image group 930 corresponding to product identifier 456 can be directly generated from 456_1.mp4, 456_2.mp4, and 456_pic1.jpg. Similarly, since the material corresponding to product identifier 789 does not exceed the object number threshold, image group 940 corresponding to product identifier 789 can be directly generated from 789_pic1.jpg.

[0178] According to embodiments of this disclosure, generating content to be published based on multiple candidate images and multiple candidate texts may include: generating creative information based on any target image from the multiple candidate images and any target text from the multiple candidate texts; and generating content to be published based on the creative information and content information corresponding to the content identifier.

[0179] A target image refers to a single candidate image selected from a set of images, and a target text refers to a single candidate text selected from a set of texts. Creative information is formed by combining the target image and the target text.

[0180] In one example, you can iterate through each candidate image in the image group as the target image and randomly select a target text from the text set, then combine the target image with the target text. For example, you can combine a video file with the creative copy "Many good products, cheap and affordable" to form a complete creative message.

[0181] Building upon this foundation, coarse-grained content data can be created based on the generated coarse-grained identifiers, such as setting budgets and ad promotion dates; fine-grained content data can be created based on the generated fine-grained identifiers, such as setting bids and bidding modes; and creative data can be created based on creative information, such as setting object identifiers and creative titles. Thus, a complete, tree-like structure of content to be published is formed based on the coarse-grained content data, fine-grained content data, and creative data.

[0182] In the embodiments disclosed herein, by automatically and randomly combining candidate objects and candidate texts and binding them to specific hierarchical structures, diversified content generation and standardized assembly are achieved, improving the efficiency of content generation and the diversity of creativity; at the same time, the structured data organization method can ensure the accuracy of content delivery, providing a reliable foundation for cross-channel batch delivery.

[0183] In this case, each candidate entity has a corresponding candidate channel identifier and a candidate number; determining whether there is a target entity for publishing the content to be published based on the target channel identifier may include: filtering multiple candidate entities based on the target channel identifier to obtain multiple intermediate entities; filtering multiple intermediate entities based on a target number threshold to obtain a filtering result; and determining the candidate entity as the target entity in response to the filtering result indicating that there is a candidate entity that satisfies the target channel identifier and the target number threshold.

[0184] Candidate channel identifiers refer to the channel identifiers to which each candidate entity belongs. Intermediate entities refer to the entities that initially meet the criteria after the first round of screening. By traversing the entity mapping, the candidate channel identifiers of each candidate entity are matched with the target channel identifiers in the configuration information, and the candidate entities that match are retained as intermediate entities.

[0185] The number of candidates refers to the number of data items created under each candidate subject. For example, the target number threshold may include a coarse-grained number threshold and a fine-grained number threshold, whereby the fine-grained number threshold may be determined based on the number of coarse-grained data items created and the allowed number of each fine-grained data item.

[0186] For each intermediate subject, the actual coarse-grained count of the created coarse-grained data (i.e., campaign_count) and the actual fine-grained count of the created fine-grained data (i.e., adgroup_count) can be retrieved from Table 5 (i.e., account_material_count). The actual coarse-grained count is compared with a coarse-grained count threshold, and the actual fine-grained count is compared with a fine-grained count threshold. If there are candidate subjects that meet both thresholds, then the screening result can be determined, indicating the existence of candidate subjects that meet the preset screening rules.

[0187] Table 5

[0188]

[0189] In the embodiments disclosed herein, a two-level screening mechanism is used to first ensure the matching of the channel and then verify the capacity limit of the subject. This achieves accurate and automated screening of the target subject, prevents content creation failure due to channel mismatch or account over-limit, ensures the smooth execution of cross-channel batch delivery, and improves the resource utilization and operation success rate of the content publishing process.

[0190] According to embodiments of this disclosure, publishing content to be published to a target entity may include: determining a content publishing component based on a target channel identifier; and publishing the content to be published to the target entity by invoking the content publishing component.

[0191] Each channel can have its own candidate publishing components. These candidate publishing components are API interfaces provided by each channel for creating content. A mapping relationship between candidate channel identifiers and candidate publishing components can be maintained in advance. Once the target channel identifier is determined, the corresponding candidate publishing component is selected and called as the content publishing component based on the mapping relationship. Then, by calling the selected content publishing component, the complete content to be published is sent to the target audience, thus achieving cross-channel content publishing.

[0192] In the embodiments of this disclosure, the automatic publishing of content on different channels is realized through the dynamic matching mechanism of channel identifier and publishing component, which solves the problem of difficulty in batch publishing content across channels and improves the coverage and operational efficiency of content publishing.

[0193] According to embodiments of this disclosure, each candidate object in the object set has a corresponding usage status; the content publishing method may further include: updating the number of candidates corresponding to the target subject; and updating the usage status of each candidate object used to generate the content to be published.

[0194] The number of candidates refers to the actual coarse-grained number and the actual fine-grained number corresponding to the target subject. For example, the `campaign_count` and `adgroup_count` fields in the table above. After content publication, the actual coarse-grained number and the actual fine-grained number corresponding to the current target subject can be updated in the table.

[0195] Usage status refers to a flag indicating whether a candidate object has been used to create content. For example, in the `use_flag` field of the table above, 0 indicates unused, 1 indicates used, 2 indicates in use, and 3 indicates prohibited. Once the content is published, the value of the field corresponding to `use_flag` in the table can be updated to 1, indicating that the candidate object has been used.

[0196] In the embodiments of this disclosure, an accurate status tracking mechanism is established by updating the usage status of candidate objects and the actual coarse-grained and fine-grained number of target subjects in real time. This prevents the repeated use of candidate objects and the over-deployment of subject resources, ensuring the accuracy of resource allocation and the reliability of the content publishing process.

[0197] Figure 10 The illustration shows an example schematic diagram of an automated content publishing process according to an embodiment of the present disclosure.

[0198] like Figure 10 As shown in the Automated Content Publishing Process 1000, the complete process from automated product selection to final content publishing is described, including the automated product selection stage, the object generation task configuration stage, the content generation task configuration stage, the object generation stage, and the content creation stage.

[0199] For the automated product selection phase, automated product selection technology is used to filter out products that meet the campaign objectives. The filtered product list is then displayed, and the selected products are grouped into a single product collection.

[0200] In the object generation task configuration phase, specific requirements for object generation can be defined, such as the quantity and specifications of videos, images, and text, and the task can be distributed to the object generation component.

[0201] In the object generation phase, the object generation component receives and processes tasks. It reads the first sub-configuration information of the task and the product information from the product collection. Based on the product information, it generates text objects and stores them in a text collection; it generates video objects and stores them in a video collection; and it generates image objects and stores them in an image collection.

[0202] In the content generation task configuration phase, specific requirements for content creation can be defined. In the content creation phase, the second-level configuration information, product information, and object information of the task can be read separately. The generated candidate objects are matched with products to ensure correct association between candidate objects and products. The content to be published is then published to the target audience.

[0203] The above are merely exemplary embodiments, but are not limited thereto. Other content publishing methods known in the art may also be included, as long as they can improve the effectiveness of content publishing.

[0204] Figure 11 A block diagram of a content publishing apparatus according to an embodiment of the present disclosure is shown schematically.

[0205] like Figure 11 As shown, the content publishing device 1100 may include a content generation module 1110, a subject determination module 1120, and a content publishing module 1130.

[0206] The content generation module 1110 is used to generate content to be published in response to a content publishing operation, based on an object set corresponding to the configuration information indicated by the content publishing operation. The object set is created based on the product set, and the configuration information indicates at least one target channel identifier and a target number threshold.

[0207] The subject determination module 1120 is used to determine whether there is a target subject for publishing content to be published based on the target channel identifier. The target subject is an account under the channel corresponding to the target channel identifier, and the number of target subjects meets the target number threshold.

[0208] The content publishing module 1130 is used to publish the content to be published to the target entity in response to the determination that there is a target entity for publishing the content to be published.

[0209] It should be noted that the content publishing device part in the embodiments of this disclosure corresponds to the content publishing method part in the embodiments of this disclosure. For a detailed description of the content publishing device part, please refer to the content publishing method part, which will not be repeated here.

[0210] Figure 12 A block diagram of an electronic device suitable for implementing a content publishing method according to an embodiment of the present disclosure is shown schematically. Figure 12The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0211] like Figure 12 As shown, a computer electronic device 1200 according to an embodiment of the present disclosure includes a processor 1201, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage portion 1209 into a random access memory (RAM) 1203. The processor 1201 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1201 may also include onboard memory for caching purposes. The processor 1201 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0212] RAM 1203 stores various programs and data required for the operation of electronic device 1200. Processor 1201, ROM 1202 and RAM 1203 are interconnected via bus 1204.

[0213] According to embodiments of this disclosure, the electronic device 1200 may further include an input / output (I / O) interface 1205, which is also connected to a bus 1204. The electronic device 1200 may also include one or more of the following components connected to the input / output (I / O) interface 1205: an input section 1206 including a keyboard, mouse, etc.; an output section 1207 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN card, modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the input / output (I / O) interface 1205 as needed. A removable medium 1211, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1210 as needed so that computer programs read from it can be installed into the storage section 1208 as needed.

[0214] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The aforementioned computer-readable storage medium carries one or more programs, which, when executed, implement the content publishing method according to the embodiments of this disclosure.

[0215] In this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0216] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this disclosure. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the content publishing methods provided in the embodiments of this disclosure.

[0217] When the computer program is executed by the processor 1201, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0218] According to embodiments of this disclosure, program code for executing computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages.

[0219] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. It should also be noted that in some alternative implementations, the functions indicated in the boxes may occur in a different order than those shown in the drawings.

[0220] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A content publishing method, comprising: In response to a content publishing operation, content to be published is generated based on a set of objects corresponding to the configuration information indicated by the content publishing operation, wherein the set of objects is created based on a set of products, and the configuration information indicates at least one target channel identifier and a target number threshold. Based on the target channel identifier, determine whether there exists a target entity for publishing the content to be published, wherein the target entity is an account under the channel corresponding to the target channel identifier, and the number of target entities meets the target number threshold; and In response to the determination that the target subject exists, the content to be published is published to the target subject.

2. The method according to claim 1, wherein, The configuration information includes first sub-configuration information and second sub-configuration information; The step of generating content to be published based on the set of objects corresponding to the configuration information of the content publishing operation instruction includes: In response to an object generation configuration operation for the first configuration interface, the object set is generated based on the product set and the first sub-configuration information corresponding to the object generation configuration operation; and In response to the content generation configuration operation for the second configuration interface, the content to be published is generated based on the object set and the second sub-configuration information corresponding to the content generation configuration operation.

3. The method according to claim 2, wherein, The product set includes multiple product information items, and the object set includes a sub-object set for each product information item. The step of generating the object set based on the product set and the first sub-configuration information corresponding to the object generation configuration operation includes: A publish-subscribe service is constructed by using the first sub-configuration information of each of the multiple product information items as the publisher and the object generation component as the subscriber; and Based on the publish-subscribe service, each of the first sub-configuration information is subscribed to, so as to generate a set of sub-objects corresponding to the product information based on the first sub-configuration information.

4. The method according to claim 3, wherein, The first configuration interface displays an object configuration control, which is used to control the object type and number of generated objects. The object type includes at least one of image type and text type, and the sub-object set includes at least one of image set and text set. The step of generating a set of sub-objects corresponding to the product information based on the first sub-configuration information includes: Based on the number of objects for the text type, generate text objects and store the text objects in the text collection; Based on the number of objects for the image type, a first candidate object is generated; and In response to the successful execution of object storage service for the candidate object, the first candidate object is stored as an image object in the image collection.

5. The method according to claim 3, further comprising: In response to a second candidate object uploaded via the object upload interface, the object name of the second candidate object is validated according to a preset validation rule to obtain a validation result. The preset validation rule defines at least one of a formatting rule for the object name and a matching rule for product identifiers in the object name. In response to the verification result indicating that the object name has passed the verification, the second candidate object is stored as an image object in the image collection.

6. The method according to claim 5, wherein, The product collection includes information on multiple products; The method further includes: The product information is displayed through a product information display interface, which includes configuration controls and object upload controls. In response to the configuration control being triggered, the first configuration interface or the second configuration interface is displayed; and In response to the object upload control being triggered, the object upload interface is displayed.

7. The method according to claim 2, wherein, The second sub-configuration information includes priority information; The step of generating the content to be published based on the object set and the second sub-configuration information corresponding to the content generation configuration operation includes: In response to the first scheduled task, the second sub-configuration information is filtered according to the execution status corresponding to the second sub-configuration information and the priority information to obtain the target sub-configuration information; and The content to be published is generated based on the object set and the target sub-configuration information.

8. The method according to claim 7, wherein, The target sub-configuration information includes product identifiers, and the object set includes an image set and a text set; The step of generating the content to be published based on the object set and target sub-configuration information includes: Based on the product identifier, obtain multiple candidate images from the image set and multiple candidate texts from the text set; and The content to be published is generated based on the multiple candidate images and the multiple candidate texts.

9. The method according to claim 8, wherein, The target sub-configuration information also includes an object number threshold; The method further includes: Based on the object number threshold, the multiple candidate images are grouped to obtain multiple image groups, wherein the number of candidate images in each image group is less than or equal to the object number threshold; and For each image group, a content identifier is generated based on the product identifier.

10. The method according to claim 9, wherein, The content identifier includes coarse-grained identifiers and fine-grained identifiers, and the product identifier corresponds to multiple levels of category names; The step of generating a content identifier based on the product identifier includes: The coarse-grained identifier is generated based on the product identifier and the category name at the previous level; as well as The fine-grained identifier is generated based on the product identifier and the category name at the next level.

11. The method according to claim 9, wherein, The step of generating the content to be published based on multiple candidate images and multiple candidate texts includes: Creative information is generated based on any target image from a plurality of candidate images and any target text from a plurality of candidate texts; and The content to be published is generated based on the creative information and the content information corresponding to the content identifier.

12. The method according to any one of claims 1 to 11, wherein, The step of determining whether there is a target entity for publishing the content to be published based on the target channel identifier includes: Based on the target channel identifier, multiple candidate entities are screened to obtain multiple intermediate entities, each of which has a corresponding candidate channel identifier and a number of candidates; Based on the target number threshold, the multiple intermediate entities are filtered to obtain the filtering results; and In response to the screening results indicating the existence of candidate entities that satisfy the target channel identifier and the target number threshold, the candidate entities are identified as the target entities.

13. The method according to claim 12, wherein, The step of publishing the content to be published to the target subject includes: Based on the target channel identifier, determine the content publishing component; and By invoking the content publishing component, the content to be published is published to the target subject.

14. The method according to claim 12, wherein, Each candidate object in the object set has a corresponding usage status; The method further includes, after publishing the content to be published to the target subject: Update the number of candidates corresponding to the target entity; as well as The usage status of each candidate object used to generate the content to be published is updated.

15. The method according to claim 1, wherein, The product collection was created in the following way: In response to the second scheduled task, the multiple candidate product information obtained from the data mapping corresponding to the content publishing operation is evaluated according to the preset evaluation index, and the evaluation values ​​of the multiple candidate product information are obtained respectively. as well as The product set is determined based on the top N candidate product information from the sorted pool of candidate product information, where N is a positive integer.

16. A content publishing device, comprising: The content generation module is used to generate content to be published in response to a content publishing operation, based on a set of objects corresponding to the configuration information indicated by the content publishing operation, wherein the set of objects is created based on a set of products, and the configuration information indicates at least one target channel identifier and a target number threshold. The subject determination module is used to determine, based on the target channel identifier, whether there exists a target subject for publishing the content to be published, wherein the target subject is an account under the channel corresponding to the target channel identifier, and the number of target subjects meets the target number threshold; and The content publishing module is used to publish the content to be published to the target entity in response to the determination that the target entity exists.

17. An electronic device comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 15.

18. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 15.

19. A computer program product comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 15.

Images