HK40124066A - Method and electronic equipment for providing social media network account operation management service - Google Patents

Description

Methods and electronic devices for providing social media network account operation and management services

Technical Field

This application relates to the field of information processing technology, and in particular to methods and electronic devices for providing social media network account operation and management services.

Background Technology

Social networking service (SNS) operations have become an indispensable part of modern corporate marketing strategies. By managing and optimizing their online accounts on social media platforms and publishing content such as posts, businesses can increase brand exposure, enhance user engagement, and boost website traffic. In today's highly competitive market environment, successful SNS operation strategies can not only win more consumer attention for businesses but also enhance their brand's competitive advantage.

The core of SNS operations lies in enhancing brand influence, increasing user engagement, and driving traffic through effective content management and user interaction. With the rapid development of social media, businesses are increasingly emphasizing SNS operations, providing ample room for innovation in related technologies and methods, and creating significant market demand.

To meet the SNS operation needs of enterprises, some products or tools can provide enterprises with comprehensive social media content management functions, such as title optimization, potential trending topic lists, rewriting of popular posts, copywriting polishing, and suggestions for the best posting time, to help enterprises improve the quality of social media interaction, or help enterprises increase the exposure and engagement of tweets, etc.

However, in the process of SNS operation, user interaction usually exhibits obvious localization characteristics. If a company or platform only targets users within a certain country, the company is usually very familiar with the culture of that country, and users are likely to be familiar with the company or platform. The company or platform's operators usually know very well which content is likely to attract user attention, etc. Therefore, it is usually feasible to combine this with the aforementioned content optimization to improve content quality. However, some companies or platforms may be involved in cross-border trade. In this case, they may need to conduct cross-border SNS operations targeting users in multiple different countries or regions. Different countries/regions have different cultures, and users in different countries/regions may have different hot topics. Users in other countries/regions may not be very familiar with specific companies/platforms, and there are also differences such as language and time zones. It is evident that the difficulty of SNS operation in cross-border scenarios is quite significant. In the current technology, companies/platforms usually need to deploy local operation personnel in different countries/regions. These operation personnel need to have a deep understanding of the local cultural background, keen insight into current events, and good content creation capabilities, etc., which makes the operation very costly.

Summary of the Invention

This application provides methods and electronic devices for providing social media network account operation and management services, which can improve the effectiveness of account operation in cross-border scenarios and reduce operating costs.

This application provides the following solution:

A method for providing social media network account operation and management services, applied to an account operation and management service system, the method comprising:

A social media network (SNS) account matrix system for the target entity is defined. This matrix system includes the target entity's main account registered on the SNS network, multiple accounts registered in different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity on the SNS network, and/or accounts registered by other entities on the SNS network and authorized to the target entity for operation and management. Each sub-account has its own account profile positioning targeting specific demographics, industries, and/or product categories.

The AI model automatically generates and publishes content according to the content publishing strategies corresponding to the accounts in the account matrix system; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning.

Among these, the process of generating content for the sub-accounts using a large AI model includes:

Collect hotspot data;

The hot data is understood by using a large AI model for decision-making to determine the appropriate content type, as well as the appropriate country/region, target audience, industry and/or product category, and to identify at least one target sub-account related to the target country/region, target audience, industry and/or product category;

The AI model, pre-trained for the target sub-account, generates content of the corresponding type for the target sub-account and determines the publication time according to the corresponding publication time strategy.

The target entities include cross-border commodity information service platforms;

The content types include product recommendation categories;

The method further includes:

The target products related to the hot data and the relevant display materials of the target products are determined from the product information service platform.

The relevant display materials for the target product are provided to the AI model trained for the target sub-account, so as to generate text content related to product recommendations based on the hot data, and assemble it with the display materials into product recommendation content to be published.

The process of generating content for the sub-account using a large AI model also includes:

After generating content of the corresponding type for the target sub-account, the generated content is adapted to the SNS network corresponding to the target sub-account in terms of content style, format, specifications, required interactive elements and/or user preferences.

During the adaptation process, if the SNS network supports multiple different content formats, the generated content will be rewritten in multiple ways to generate content corresponding to multiple content formats, and then published in different content formats respectively.

This also includes:

The content publishing strategy for the sub-accounts will be updated based on user feedback regarding the content posted on the accounts.

This also includes:

The AI-powered large-scale model enables collaborative interaction among different accounts within the account matrix system.

The collaborative interaction processing between different accounts in the account matrix system includes:

The quality of the content generated by the AI model for the sub-account is judged. If it meets the preset conditions, the generated content mentions the country/region account of the sub-account and/or the main account, so as to realize the interaction between the sub-account and the country/region account and/or the main account.

The collaborative interaction processing between different accounts in the account matrix system includes:

The sub-accounts are used to forward some of the content posted by the country/region accounts of their respective countries/regions.

The collaborative interaction processing between different accounts in the account matrix system includes:

Sub-accounts with the same or related account profiles follow each other and/or forward each other's content.

This also includes:

After receiving comments or private messages from users regarding published content, accounts in the account matrix system use AI big data models to generate responses to the comments or private messages.

This also includes:

The system uses sub-accounts within the account matrix to follow well-known accounts outside the corresponding industry or field, and after these well-known accounts publish content, it uses an AI model to generate comments on the published content.

This also includes:

By setting up keyword and/or topic monitoring mechanisms, we can capture and analyze trending topics and/or discussion trends in SNS networks in real time, identify sub-accounts in the account matrix system that are related to the trending topics and/or discussion trends, and generate comments on the trending topics and/or discussion trends by the sub-accounts through an AI big data model.

A method for processing SNS account information includes:

The target interface receives account configuration information submitted by the target entity. The account configuration information includes the main account registered by the target entity in the SNS network, multiple accounts registered in multiple different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity in the SNS network, and/or accounts registered by other entities in the SNS network and authorized to the target entity for operation and management. The sub-accounts have their own account profile positioning in terms of targeting specific sub-segments of people, industries and/or product categories.

Based on the account configuration information, an SNS account matrix system for the target entity is constructed so that content can be automatically generated and published by an AI big data model according to the content publishing strategies corresponding to the accounts in the account matrix system; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning.

The AI model is also used for collaborative interaction between different accounts.

A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of any of the preceding methods.

An electronic device, comprising:

One or more processors; and

A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the preceding descriptions.

A computer program product includes a computer program/computer executable instructions that, when executed by a processor in an electronic device, implement the steps of any of the preceding methods.

According to the specific embodiments provided in this application, the following technical effects are disclosed:

This application provides account operation and management services for entities operating cross-border accounts. This service allows for the construction of an SNS account matrix system for specific entities. The matrix can include the main account registered by the entity on the SNS network, as well as country/region accounts registered for multiple countries/regions. Each country/region account can have at least one sub-account. These sub-accounts can be accounts directly registered by the current entity on the SNS network, or accounts registered by other entities on the SNS network and authorized for operation and management by the current entity. These sub-accounts have their own account profile positioning targeting specific sub-segments, industries, and/or product categories. Based on the constructed account matrix system, an AI model can be used to automatically generate and publish content according to the content publishing strategies corresponding to the accounts in the account matrix system. The content publishing strategy corresponding to a sub-account can be related to the account profile positioning, ensuring that the content, timing, etc., published by a specific sub-account are related to its specific account profile positioning. This achieves refined operation for each user group, allowing the same sub-account to gain continuous attention from its corresponding audience.

In optional implementations, AI-powered large-scale models can also facilitate collaborative interaction between different accounts within the account matrix system. This includes mentioning the country/region account and/or main account of the sub-account in the content generated for the sub-account; or, sub-accounts forwarding portions of content posted by the country/region account of their respective country/region; or, sub-accounts with the same or related account profiles mutually following each other and/or forwarding each other's content, etc. This interaction between accounts can create a broad brand image presentation for users, reducing account operation costs.

Additionally, intelligent replies can be achieved through AI big data models, including replying to user comments or private messages, or intelligently replying to content posted by popular accounts in the industry. Alternatively, by setting up keyword and/or topic monitoring mechanisms, hot topics and/or discussion trends in SNS networks can be captured and analyzed in real time, and sub-accounts related to the hot topics and/or discussion trends in the account matrix system can be identified. The AI big data model can then generate comments on the hot topics and/or discussion trends made by the sub-accounts.

Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time.

Attached Figure Description

To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

Figure 1 is a schematic diagram of the system architecture provided in an embodiment of this application;

Figure 2 is a flowchart of the first method provided in an embodiment of this application;

Figure 3 is a schematic diagram of the account matrix system provided in an embodiment of this application;

Figure 4 is a flowchart of the second method provided in an embodiment of this application;

Figure 5 is a schematic diagram of an electronic device provided in an embodiment of this application.

Detailed Implementation

The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

First, it's important to note that in cross-border trade, both platforms (those providing product information services, often referred to as e-commerce platforms) and individual merchants face the challenge that overseas users may not be familiar with the platform or the company. If a platform needs to launch large-scale marketing campaigns, or a company needs to enhance its brand awareness, leveraging social media accounts for promotion is crucial. This allows information to effectively reach consumers, fostering greater user interaction and ultimately driving traffic to the platform or specific brand stores. Currently, achieving this goal typically involves optimizing the quality of published content to attract users. However, due to significant cultural differences across countries, implementing differentiated account operations in multiple countries to achieve promotional objectives often incurs substantial operational costs.

In this embodiment, to assist relevant platforms or enterprises in managing their SNS accounts in cross-border trade scenarios, as shown in Figure 1, an account operation management service system can be provided. This system offers SNS operation management services to specific entities (e.g., e-commerce platforms or enterprises), helping them manage their SNS accounts in cross-border commodity trade scenarios. Specifically, this service can first create an SNS account matrix system for specific platforms or enterprises, establishing a three-tiered account matrix operation system of "main account - country account - sub-account." This allows for more refined operations based on country dimensions. Sub-accounts can also have their own user profiles, such as targeting specific demographics, industries, or product categories. This allows them to reach a more vertical audience and expand their reach to followers (commonly known as "fans"). In other words, these sub-accounts allow for refined operations targeting specific demographics. A sub-account can only publish content that users in that specific demographic want to see, instead of publishing indiscriminately. This allows each sub-account to maintain a specific group of followers (otherwise, if an account publishes all kinds of content, different groups of people may like different content from the same account, and some of them may reduce their attention to the account because they may feel that the account cannot be the source of information they need).

Taking a specific e-commerce platform as an example, in the aforementioned three-tier account matrix operation system, the platform's main SNS account can first be identified. Then, the same platform can be associated with multiple country/region accounts for content publishing targeting users in specific countries/regions. Furthermore, multiple sub-accounts can be associated under each specific country/region account. These sub-accounts can be multiple "small accounts" registered by the platform itself, or, in this embodiment, SNS accounts registered by merchants but authorized to the platform for operation, and so on. In other words, sub-accounts can be SNS accounts registered by the current entity itself, or SNS accounts registered by other entities but authorized to the current entity. Whether registered by the current entity itself or registered by other entities and authorized to the current entity for operation, these sub-accounts can have their own user profile positioning. That is, a specific sub-account can target a specific segment of a vertical market, a specific industry, and/or a specific product category. Correspondingly, the content published by these sub-accounts can also correspond to their user profile positioning.

Since there may be multiple social networking sites (SNS), a single entity can have multiple main accounts, each corresponding to a different SNS network. Similarly, a single entity can have multiple intermediate-level accounts within the same country/region, each corresponding to a different SNS network. Sub-accounts operate similarly, targeting the same niche audience, industry, or product category, and can each correspond to different sub-accounts within different SNS networks. This results in a massive number of accounts for the same entity within this three-tiered account matrix operation system. Furthermore, to achieve more refined operations, the content and timing of posts from different accounts may need to be different. Therefore, if the current technology were to rely on dedicated personnel to edit the content for each account, the workload would be enormous, potentially an impossible task.

Therefore, in this embodiment, based on the aforementioned three-tier account matrix system, a scheme for automated operation of the accounts based on an AI (Artificial Intelligence) large-scale model is proposed. The so-called AI large-scale model typically refers to a deep learning model containing massive amounts of parameters. Due to its large scale, this AI large-scale model can store and process large amounts of information, thereby achieving higher performance on various tasks. Therefore, the intelligent account operation in this embodiment can be achieved by relying on the powerful natural language text understanding, logical thinking, and content generation capabilities of the pre-trained AI large-scale model. Specifically, the AI large-scale model can automatically generate and publish content according to the content publishing strategy corresponding to a specific account. This content publishing strategy can be related to the account profile positioning of the specific account; that is, it is necessary to generate content suitable for different profile positioning for different accounts. It can also automatically process the publishing time, language, etc., based on the country/region to which the specific account belongs. It's important to note that in practice, the AI model primarily generates content for sub-accounts, while content published by the main account, country/region accounts, etc., usually carries a certain degree of authority or represents the image of a specific platform or brand in a particular country/region. Therefore, the published content needs to be more cautious. Furthermore, the number of main accounts and country/region accounts is not particularly large. Thus, they can be generated manually, or their AI-generated content can be manually reviewed before publication, and so on.

Furthermore, AI-powered large-scale models can be used to manage the collaborative relationships between different accounts, enabling cross-promotion and other similar features. For example, after the AI model generates specific content for a sub-account, if the content is of high quality, it can be used to mention specific country/region accounts or the main account, thereby driving traffic to those accounts. Alternatively, after the main account or country/region account publishes content, sub-accounts can repost it, allowing the content to reach a specific demographic more effectively through these sub-accounts.

In practical applications, in addition to intelligently generating and publishing specific content through AI big data models, intelligent reply content can also be generated through AI big data models. For example, when a user comments on content published by an account, reply content can be intelligently generated through AI big data models. Or, when a user sends a private message to an account, reply content can also be intelligently generated through AI big data models, and so on.

The specific implementation schemes provided in the embodiments of this application will be described in detail below.

Example 1

First, Embodiment 1 of this application provides a method for providing social media network account operation and management services. This method is applied to an account operation and management service system. Specifically, as shown in Figure 2, the method may include:

S201: Determine the social media network (SNS) account matrix system of the target entity. The account matrix system includes the main account registered by the target entity in the SNS network, multiple accounts registered in multiple different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity in the SNS network, and/or accounts registered by other entities in the SNS network and authorized to the target entity for operation and management. The sub-accounts have their own account profile positioning in terms of targeting specific sub-segments, industries, and/or product categories.

When providing account operation services through the solutions provided in this application, an SNS account matrix system can first be built for specific target entities. These specific target entities are those with account operation needs; they could be an e-commerce platform, a merchant within the e-commerce platform, or other businesses, etc.

Regarding the SNS account matrix system, it can specifically be the three-tiered account matrix system mentioned above, as shown in Figure 3. The first tier can be the main account, the middle tier can be country/region accounts, and the third tier consists of sub-accounts. These can be accounts registered by the current entity itself, or accounts registered by other entities and authorized for operation by the current entity. Of course, whether registered by the current entity or by other entities, these sub-accounts can have their own account profiles and positioning, such as targeting a specific segment of the audience in a vertical field, a certain industry, or a certain product category.

Since there are typically multiple social networking sites (SNS), a single entity can have multiple main accounts corresponding to different SNS networks. Furthermore, a single entity can also open country/region accounts on multiple different SNS networks within the same country/region. For example, assuming there are M SNS networks and N countries/regions, one entity can have a maximum of M×N country/region accounts. Of course, in practice, it's also possible to choose to open country/region accounts only on certain SNS networks and in certain countries/regions, and so on. Regarding sub-accounts, each country/region account can have multiple different sub-accounts to enable differentiated and targeted operations for different demographics within the same country/region.

In practice, this can be implemented by providing users (such as operators of a specific entity) with an interface for configuring account information for the target entity. Users can prepare various accounts in advance and enter information about these accounts through the interface, such as the main account of a target entity, the country/region accounts for each country/region, and the sub-accounts under each country/region account.

S202: Automated content generation and publishing are performed using an artificial intelligence (AI) big data model according to the content publishing strategies corresponding to the accounts in the account matrix system; wherein, the content publishing strategy corresponding to the authorized sub-account is related to the account profile positioning.

After constructing a three-tiered account matrix system, the sheer number of accounts within the matrix, and the potential for different accounts to have distinct profiles and positioning (e.g., targeting different countries/regions, specific demographics, industries, or categories), necessitates generating different content for each account to align with its specific profile when publishing content. Based on this, this application provides an implementation method for generating and publishing content for specific accounts using a large-scale AI model.

Specifically, when generating content for the authorized sub-accounts using an AI-powered large-scale model, the first step is to collect trending data, which is then used to generate specific content. This data collection can be achieved in various ways. For example, if the target is an e-commerce platform, it can collect both internal and external trending data from that platform. Internal trending data, specifically from the e-commerce platform itself, can include user search terms, new and popular products, and hot-selling events (such as large-scale promotional activities). It can also involve selecting high-performing creatives from external submissions. External trending data can be collected by integrating with third-party tools, such as trending posts on social media, best-selling lists from other e-commerce platforms, and holiday information.

In the process of capturing real-time trending data, simply generating content based on current trending data may lead to a series of problems. For example, negative trending information may not be suitable for posting, similar trending topics may cause fatigue, and the trending topic may not align with the current subject's positioning. These problems may cause users to lose interest in the content, thereby affecting the platform's overall effectiveness.

Therefore, in practical applications, after collecting hotspot data from both inside and outside the site, some data cleaning processing can be performed, which can be mainly divided into the following aspects:

Filtering of negative trending data: If trending data contains negative sentiment (such as negative comments, crisis events, etc.), the trending data can be discarded, the evaluation results can be recorded, and subsequent analysis can be carried out.

Hot Topic Fatigue Detection: Check whether hot topic data information is highly duplicated with recently published content. If so, you can abandon the publication of the content corresponding to the hot topic data, or consider differentiating the content.

Account matching filtering: Compare the current trending data with the account profile positioning associated with the current subject. If the content is significantly inconsistent with the account profile positioning (such as target audience, brand tone, etc.), the content can be abandoned.

Risk control tagging and filtering: The collected hot data can be automatically tagged for risk control.

After cleaning and processing the hot data, specific content is generated. To improve the quality of generated content, multiple content types can be predefined, and each type can be provided with detailed meanings and publishing purposes. This information can serve as prompts for the large model, enabling it to generate content more directionally and purposefully, thereby improving the quality of the generated content.

For example, when the target entity is a cross-border e-commerce platform, the specific content types and their detailed meanings could include:

Product Recommendations: Based on trending topics, products related to those topics can be recommended according to user needs and interests. For example, if the current hot topic is environmental protection, sustainable development products such as reusable bags, goods made from renewable materials, or green cleaning products can be recommended.

Event Sweepstakes: By leveraging current trending topics, a related sweepstakes event can be designed to attract user participation and enhance interactivity. For example, if the recent trending topic is the release of a popular movie, merchandise related to the movie theme can be offered as prizes.

Brand Promotion: When promoting a brand, trending events can be leveraged to enhance brand awareness and reputation. For example, when the public is paying attention to a public welfare activity, the brand can combine its social responsibility with promotional content about participating in the activity to showcase its values and sense of mission.

Hot Post Rewriting: This involves rewriting or summarizing popular posts based on current trending topics to attract user attention and participation. For example, if a TV series has recently generated a lot of discussion, the hot post could be rewritten as "Who is your favorite character?" and users could be encouraged to share their opinions and experiences.

Creative Voting: Launch creative voting campaigns around trending topics, allowing users to participate in selections or votes. For example, during a popular sporting event, a poll could be conducted asking "Who do you think will win the championship?", or a "Best Creative Costume" poll could be organized during popular holidays such as Halloween to increase user interaction and engagement.

Industry Trends: We combine trending industry news with updates on key events and trends to provide users with timely information. For example, if a popular technology product is recently released, we can write related industry analysis articles to interpret the product's impact on the market.

Knowledge Sharing: Share professional knowledge on trending topics to provide users with valuable content. For example, if the current trend is healthy eating, you can write a guide on how to choose healthy foods, or organize online lectures and invite experts to share relevant knowledge to enhance user trust and brand image.

Of course, in practice, other content types can also be included. In addition, different content types can be defined for different types of subjects, which will not be listed here.

With multiple content types predefined, a large AI model can be provided for decision-making. This model can understand collected trending data to determine suitable content types, as well as appropriate countries/regions, target audiences, industries, and/or product categories. It can also identify at least one target sub-account related to that target country/region, target audience, industry, and/or product category. In other words, for collected trending data, the AI model can analyze and determine the appropriate content type, country/region, and target audience for publication, and subsequently, which sub-accounts to use for publication. For example, assuming a trending data point is related to the "nail salon" industry and originates from a top-selling list on an e-commerce platform in a certain country/region, it can be determined that this data is suitable for publication in that country/region targeting a "nail salon" related audience. Therefore, sub-accounts related to that industry can be selected within that country/region, and so on.

After determining which sub-accounts to publish on, corresponding content can be generated. Since different sub-accounts have their own unique brand identity and positioning, even those targeting the same audience or industry may have distinct styles, tones, and content expression habits. Therefore, to diversify the generated content and reflect the characteristics of each sub-account, a separate AI model can be pre-trained for each specific sub-account. (For example, this can be achieved using a baseline model plus multiple LoRA (Low-Rank Adaptation of Large Language Models) models, with each LoRA model adapting to its respective sub-account.) In this scenario, since trending data has already been collected, content types have been determined, and the sub-accounts to publish on have been identified, the pre-trained AI models for these target sub-accounts can generate corresponding content. The publishing time can also be determined according to a specific publishing strategy. Furthermore, since specific sub-accounts often correspond to specific countries/regions, the generated content can be written in the language of that country/region.

Specifically, in the content generation process using the AI model, content publishing strategies can be tailored to individual sub-accounts. These strategies define suitable content styles and relevance for each sub-account, and the AI model generates content accordingly. However, in the early stages of a project, it may be difficult to precisely determine the appropriate style for each sub-account. Therefore, the goal is to generate as much diverse content as possible and then adjust the publishing strategy based on user feedback during the publishing process. For example, the following features may require adjustment and experimentation:

Hotspot data quality: Collect and connect to as many different data sources as possible, and mark the source channel and timeliness of hotspots as basic characteristics;

Content Types: Generate as many content templates as possible, such as product recommendation posts, raffles, brand promotions, rewriting of popular posts, polls, etc. For each type, prepare multiple Prompt templates and mark them as corresponding features;

Product Quality: When product recommendations are required in the generated content, in order to prevent quality problems caused by the massive number of products on the site, industry operations can first select various types of high-quality product pools, such as new product pools, best-selling pools, and industry pools.

The generated content can be ranked based on manual scoring by SNS operators. Through statistical learning, the basic weights of each feature are obtained. In the subsequent intelligent posting plan, the content quality score is calculated and ranked according to these weights, thus initially forming a content publishing strategy suitable for the account.

Specifically, during the content generation process, the characteristics of the target entity can be considered, and content related to that entity can be added to the content. For example, if the target entity is a cross-border commodity information service platform, as mentioned earlier, one type of content could be product recommendations. In this case, content related to the specific recommended products can be added to the generated content. Specifically, after collecting trending data and determining the product recommendation content to be generated, target products related to the trending data and related display materials for the target products, including product images, titles, and attributes, can be identified from the commodity information service platform. Then, these display materials can be provided to the AI model trained for the target sub-account, enabling the AI model to generate text content related to product recommendations based on the trending data and assemble it with the display materials into published product recommendation content.

Furthermore, when generating content for the sub-accounts using the AI big data model, the generated content can be adapted to the SNS network corresponding to the target sub-account in terms of content style, format, specifications, required interactive elements, and/or user preferences. This includes not only adjusting the tone and style of the text content but also image size, format, and interactive elements required by different platforms, such as polls and Q&A, ensuring that the final published content effectively attracts user attention and participation across various SNS channels. For example, if an SNS network supports the use of tags to make published content easier for users to discover, it provides multiple ways to reach a larger audience and increase interaction. In this case, a set of currently popular tag libraries can be provided, and the big data model can be used to find the tags that best fit the generated content and integrate them into the content.

In one scenario, assuming a social networking site requires all published content to include images, the content generation process can simultaneously acquire relevant image materials while generating text content, assembling them into the final content to be published. The images can be retrieved from relevant image libraries (such as product image libraries on e-commerce platforms), or they can be directly generated by a large AI model, and so on.

Furthermore, some social networking sites (SNS) may support multiple content formats. For example, one SNS might allow users to post "works" and also "daily life" posts. "Works" typically have a clear creative purpose and form, aiming to express the author's emotions, thoughts, and values, emphasizing artistry and aesthetics. "Works" often require deep thought and rich imagination, employing detailed and vivid descriptions and metaphors, and focusing on language, rhythm, and other techniques. "Daily life," on the other hand, focuses more on the trivialities and ordinary events of life, recording the minutiae of daily life, without emphasizing artistry or aesthetics. Daily records can be spontaneous and impromptu, not necessarily with a clear creative purpose, but more about recording real scenes and feelings in life. Additionally, the display duration differs. Posted "works" are usually permanent unless intentionally deleted, while posted "daily life" posts typically exist for a certain period (e.g., 24 hours) before being hidden, visible only to the author. The target audience also differs: "works" are visible to all users, while "daily life" posts are usually only viewable by "friends" who have followed the account and accessed the account's profile. Based on the above, when adapting the generated content for SNS networks, it can be rewritten in various ways to create content in different formats, which can then be published in different formats. For example, content suitable for publication as a work of art can be generated, as well as content suitable for everyday publication, and so on. Alternatively, after generating content suitable for publication as a work of art, the current content can be modified through a larger model to conform to an "everyday" style, including shortening the text length to make it more immediate and visually impactful.

It should be noted that when posting through a product information service system, it may be necessary to frequently add specific product information to the content. However, if product information is added directly to the content in the form of an artwork, there may be issues such as the frequency being limited, otherwise it may be restricted by the rules of the SNS platform. In this case, you can also choose to rewrite the "artwork" content as "daily" content, and then associate specific product information in the "daily" content, and so on.

During the content publishing phase, a series of steps can be followed to ensure the safety and effectiveness of the content. First, anti-AI detection is performed to determine if the generated content meets the platform's security standards. Next, risk control verification is conducted to ensure that the published content will not cause legal or ethical issues, thereby reducing risk. Then, the potential performance of the content can be evaluated based on AI scoring. Based on this, a detailed publishing plan can be developed using AI, selecting the optimal publishing time to maximize the content's exposure and impact. Finally, after human scoring to reconfirm the content's quality and dissemination potential, it can be officially published to various SNS platforms, ensuring the content successfully reaches the target audience. This systematic publishing process can effectively enhance the value of the content and the user's interactive experience.

In addition, during the campaign phase, user behavior data on the platform can be directly fed back (such as views, interactions, likes, etc.) to further guide and improve posting strategies and quality, including:

(1) Optimization strategy model based on EE (Exploration & Exploitation, where exploration is developing unknown areas and exploitation is utilizing known states).

Test different content publishing strategies, including different content styles, types, timing, and frequencies. At this stage, focus on the diversity of user behavior to identify underutilized content formats or publishing times. Based on previous user feedback and data analysis, determine the best-performing content styles and types, and reuse and optimize them extensively to maximize user interaction and engagement.

(2) Optimization of release timing and frequency

By analyzing user activity periods, we can understand when to post to achieve the best interaction. At the same time, we test different posting frequencies to ensure we find the optimal time and frequency for posting without unduly disturbing users.

(3) Optimization of copywriting quality

Based on user feedback, similar high-quality content from the same account can be collected and used to continuously train the AI posting copy generation model for the corresponding strategy. The improvement effect can be confirmed through methods such as manual scoring and A/B testing.

(4) Product quality optimization

When the product information service system is the main platform for posting, in addition to estimating product quality through product model scoring within the site, users can also be asked to provide feedback on their preferences for products by launching questionnaires, and products in the product pool can be directly replaced based on the voting data.

(5) Improved account profile

Continuously monitoring and analyzing user browsing and interaction behaviors, and combining this with user data feedback from the platform, will continuously improve the user preferences and profiles of the corresponding user groups for each account, thereby driving the iterative upgrade of content production quality and strategies.

In optional implementations, besides generating post content through an AI model, it's also possible to manage collaborative relationships between different accounts to achieve mutual traffic redirection. Specifically, in one approach, the quality of content generated by the AI model for the sub-account can be assessed. If it meets preset conditions, the generated content can mention the country/region account and/or the main account of the sub-account's country/region, enabling interaction between the sub-account and the country/region account and/or the main account. In other words, after the AI model generates post content for a sub-account, if the post content is of good quality, it can mention the specific country/region account using methods such as "@country/region account." For example, if a sub-account posts content related to a product, it can include a message like, "For more details about this product, please visit @country/region account," etc. If users are interested, they can click the country/region account link in the content to access the country/region account's homepage, thus redirecting traffic to that country/region account, and so on.

In another approach, when managing collaboration between different accounts, sub-accounts can be used to forward content posted by accounts from their respective countries/regions. For example, if an account from a certain country/region posts content about a platform activity, a sub-account associated with that country/region can forward that content. This allows the content posted by the country/region account to reach specific niche audiences in certain vertical fields more precisely through sub-accounts.

Furthermore, it also enables "mutual following and sharing" between different sub-accounts, allowing content posted by one sub-account to reach a wider audience through reposting by other sub-accounts. Of course, during the "mutual following and sharing" process between sub-accounts, it's best to avoid content sharing between sub-accounts from different industries to prevent impacting user experience.

The above mainly describes the solution provided in this application embodiment from the perspective of intelligent posting. In practical applications, in addition to intelligent posting, intelligent replies can also be achieved through AI large-scale models. For example, after an account posts some content, some users may comment on the content, and may ask questions in the comments, such as how to buy a certain product, etc. In addition, some users may also send private messages to the account after viewing the content posted by an account to ask certain questions. In the case of numerous sub-accounts in this application embodiment, the user comments or private messages may be very numerous. Therefore, after the accounts in the account matrix system receive the comments or private messages input by users regarding the posted content, they can also use AI large-scale models to generate reply content to the comments or private messages.

In addition to replying to user comments or private messages to accounts within the system, the system also allows for mutual following and sharing between accounts. Therefore, it can intelligently reply to followed accounts after they post content. For example, a main account with sub-accounts A and B in a specific country/region might follow each other due to their related industries. After sub-account A posts content, the AI model can generate a reply in the voice of sub-account B, and so on.

Furthermore, accounts within the system can also follow popular accounts outside the system, such as influential accounts in a certain industry. By following these popular accounts, accounts within the system can increase their exposure by commenting on the content posted by these popular accounts in a timely manner.

Alternatively, besides content from well-known and popular accounts that may generate significant buzz, content from lesser-known accounts can also unexpectedly become trending. Comments on such content can be promptly posted through accounts within the system. To achieve this, keyword and/or topic monitoring mechanisms can be implemented to capture and analyze trending topics and discussion trends on social networking platforms in real time. Using AI-powered big data model contextual understanding technology, specific sub-accounts within the account matrix system related to the topic or discussion trend can be identified, and highly relevant responses can be generated, guiding users to those sub-accounts for deeper interaction and understanding. For example, in the nail industry, there are several popular keywords. Monitoring can be conducted to see if any posts are trending under a particular keyword (e.g., a surge in likes). If so, AI-powered big data models can be used to generate comments on these trending posts through relevant industry sub-accounts within the system, and so on.

Specifically, during the intelligent reply process, the quality of intelligent replies can be evaluated to gradually improve it. However, while the quality of content can be directly assessed from user browsing and interaction behaviors during intelligent posting, this data is not readily available for replies. Therefore, the core solution is to use a large AI model to categorize user intent, establish different user behavior feedback standards for different categories, and then iterate and optimize them step by step. For example, the user reply type distribution of a certain cross-border commodity information service system is as follows:

Inquiries/Complaints: 50%

Emotional venting: 30%

Sharing category: 10%

Taking consultation as an example, after replying to user inquiries, feedback on the reply can be obtained through private messages or other means. The adoption rate is calculated based on this feedback and used to tag the AI-generated replies, thereby driving continuous upgrades to the AI model and prompt strategy to improve the quality of replies.

For emotional outbursts, AI can generate soothing responses, and the success of the response strategy can be judged by observing whether the user complains again.

It should be noted that the AI big model used in this application embodiment can be based on existing content generation capabilities such as existing big language models. After cleaning and inputting information such as internal and external data and script templates, the model is trained to meet the needs of specific subject sites, so that it can generate content that conforms to the characteristics and brand tone of different platforms.

In summary, this application provides account operation and management services for entities operating cross-border accounts. This service allows for the construction of an SNS account matrix system for specific entities. The matrix can include the main account registered by the entity on the SNS network, as well as country/region accounts registered for multiple countries/regions. Each country/region account can have at least one sub-account. These sub-accounts can be accounts directly registered by the current entity on the SNS network, or accounts registered by other entities on the SNS network and authorized for operation and management by the current entity. These sub-accounts have their own account profile positioning targeting specific sub-segments, industries, and/or product categories. Based on the constructed account matrix system, an AI model can be used to automatically generate and publish content according to the content publishing strategies corresponding to the accounts in the account matrix system. The content publishing strategy corresponding to a sub-account can be related to the account profile positioning, ensuring that the content and timing published by a specific sub-account are related to its specific account profile positioning. This achieves refined operation for each user group, allowing the same sub-account to gain continuous attention from its corresponding audience.

In optional implementations, AI-powered large-scale models can also facilitate collaborative interaction between different accounts within the account matrix system. This includes mentioning the country/region account and/or main account of the sub-account in the content generated for the sub-account; or, sub-accounts forwarding portions of content posted by the country/region account of their respective country/region; or, sub-accounts with the same or related account profiles mutually following each other and/or forwarding each other's content, etc. This interaction between accounts can create a broad brand image presentation for users, reducing account operation costs.

Additionally, intelligent replies can be achieved through AI big data models, including replying to user comments or private messages, or intelligently replying to content posted by popular accounts in the industry. Alternatively, by setting up keyword and/or topic monitoring mechanisms, hot topics and/or discussion trends in SNS networks can be captured and analyzed in real time, and sub-accounts related to the hot topics and/or discussion trends in the account matrix system can be identified. The AI big data model can then generate comments on the hot topics and/or discussion trends made by the sub-accounts.

Example 2

Embodiment 2 corresponds to Embodiment 1 above. From the user's perspective (e.g., the operator of the target entity), it provides a method for processing SNS account information. Referring to Figure 4, the method may include:

S401: Receive account configuration information submitted by the target entity through the target interface. The account configuration information includes the main account registered by the target entity in the SNS network, multiple accounts registered for multiple different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity in the SNS network, and/or accounts registered by other entities in the SNS network and authorized to the target entity for operation and management. The sub-accounts have their own account profile positioning in terms of targeting specific sub-segments of people, industries, and/or product categories.

S402: Construct an SNS account matrix system for the target entity based on the account configuration information, so as to automatically generate and publish content according to the content publishing strategy corresponding to the account in the account matrix system through an AI big data model; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning.

The AI model can also be used for collaborative interaction between different accounts.

For the parts of this embodiment that are not described in detail, please refer to the description in embodiment one and other parts of this specification, which will not be repeated here.

It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., with the user's explicit consent, with the user being properly notified, etc.).

Corresponding to Embodiment 1, this application also provides an apparatus for providing social media network account operation and management services. This apparatus is applied to an account operation and management service system and includes:

The account matrix system determination unit is used to determine the social media network (SNS) account matrix system of a target entity. The account matrix system includes the target entity's main account registered on the SNS network, multiple accounts registered in different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity on the SNS network, and/or accounts registered by other entities on the SNS network and authorized to the target entity for operation and management. Each sub-account has its own account profile positioning targeting specific demographics, industries, and/or product categories.

The account management service unit is used to automatically generate and publish content according to the content publishing strategy corresponding to the account in the account matrix system using an artificial intelligence (AI) big data model; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning.

Specifically, when the account management service unit generates content for the sub-account using the AI big data model, it can be used for:

Collect hotspot data;

The hot data is understood by using a large AI model for decision-making to determine the appropriate content type, as well as the appropriate country/region, target audience, industry and/or product category, and to identify at least one target sub-account related to the target country/region, target audience, industry and/or product category;

The AI model, pre-trained for the target sub-account, generates content of the corresponding type for the target sub-account and determines the publication time according to the corresponding publication time strategy.

Specifically, the target entities include cross-border commodity information service platforms; the content types include product recommendation categories.

At this point, the device may also include:

The display material acquisition unit is used to determine, from the commodity information service platform, target commodities related to the hot data, and related display materials of the target commodities;

The display material providing unit is used to provide relevant display materials of the target product to the AI big model trained for the target sub-account, so as to generate text content related to product recommendations based on the hot data, and assemble it with the display materials into product recommendation content to be published.

In addition, the account management service unit, when generating content for the sub-account using the AI big data model, can also be used for:

After generating content of the corresponding type for the target sub-account, the generated content is adapted to the SNS network corresponding to the target sub-account in terms of content style, format, specifications, required interactive elements and/or user preferences.

Additionally, it can also be used to: when performing the adaptation process, if the SNS network supports multiple different content formats, rewrite the generated content in multiple ways to generate content corresponding to multiple content formats, and publish them in different content formats respectively.

Furthermore, the account management service unit can also be used for:

The content publishing strategy for the sub-accounts will be updated based on user feedback regarding the content posted on the accounts.

Specifically, the account management service unit is also used for:

The AI-powered large-scale model enables collaborative interaction among different accounts within the account matrix system.

Specifically, when handling collaborative interactions between different accounts within an account matrix system, it can be used for:

The quality of the content generated by the AI model for the sub-account is judged. If it meets the preset conditions, the generated content mentions the country/region account of the sub-account and/or the main account, so as to realize the interaction between the sub-account and the country/region account and/or the main account.

Alternatively, the sub-account can be used to forward some of the content posted by the country/region account of the country/region to which it belongs.

Alternatively, sub-accounts with the same or related account profiles can follow each other and/or forward each other's content.

Additionally, the device may also include:

An automatic reply unit is used to generate a reply to a comment or private message from a user after the account in the account matrix system receives such a comment or private message from the user regarding the posted content, using an AI big data model.

The comment content generation unit is used to follow well-known accounts outside the corresponding industry or field through sub-accounts in the account matrix system, and generate comment content for the published content using an AI big data model after the well-known accounts outside the system publish content.

Alternatively, by setting up keyword and/or topic monitoring mechanisms, hot topics and/or discussion trends in SNS networks can be captured and analyzed in real time, and sub-accounts related to the hot topics and/or discussion trends in the account matrix system can be identified. AI big data models can then be used to generate comments on the hot topics and/or discussion trends made by these sub-accounts.

Corresponding to Embodiment 2, this application also provides an SNS account information processing device, which may include:

The configuration information receiving unit is used to receive account configuration information submitted by the target entity through the target interface. The account configuration information includes the main account registered by the target entity in the SNS network, multiple accounts registered in multiple different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity in the SNS network, and/or accounts registered by other entities in the SNS network and authorized to the target entity for operation and management. The sub-accounts have their own account profile positioning in terms of targeting specific sub-segments, industries, and/or product categories.

The account system construction unit is used to construct an SNS account matrix system for the target subject based on the account configuration information, so as to automatically generate and publish content according to the content publishing strategy corresponding to the account in the account matrix system through an AI big data model; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning.

The AI model can also be used for collaborative interaction between different accounts.

In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

And an electronic device, comprising:

One or more processors; and

A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

A computer program product includes a computer program/computer executable instructions that, when executed by a processor in an electronic device, implement the steps of the method described in the foregoing method embodiments.

Figure 5 illustrates the architecture of an electronic device, which may include a processor 510, a video display adapter 511, a disk drive 512, an input/output interface 513, a network interface 514, and a memory 520. The processor 510, video display adapter 511, disk drive 512, input/output interface 513, network interface 514, and memory 520 can communicate with each other via a communication bus 530.

The processor 510 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

The memory 520 can be implemented in the form of ROM (Read-Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 520 can store the operating system 521 for controlling the operation of the electronic device 500, and the basic input/output system (BIOS) for controlling the low-level operations of the electronic device 500. Additionally, it can store a web browser 523, a data storage management system 524, and an account information processing system 525, etc. The aforementioned account information processing system 525 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 520 and is called and executed by the processor 510.

Input/output interface 513 is used to connect input/output modules to realize information input and output. Input/output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

Network interface 514 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

Bus 530 includes a pathway for transmitting information between various components of the device, such as processor 510, video display adapter 511, disk drive 512, input/output interface 513, network interface 514, and memory 520.

It should be noted that although the above-described device only shows the processor 510, video display adapter 511, disk drive 512, input/output interface 513, network interface 514, memory 520, bus 530, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

The foregoing has provided a detailed description of the method and electronic device for providing social media network account operation and management services. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims (18)

1. A method for providing social media network account operation and management services, characterized in that it is applied to an account operation and management service system, the method comprising: A social media network (SNS) account matrix system for the target entity is defined. This matrix system includes the target entity's main account registered on the SNS network, multiple accounts registered in different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity on the SNS network, and/or accounts registered by other entities on the SNS network and authorized to the target entity for operation and management. Each sub-account has its own account profile positioning targeting specific demographics, industries, and/or product categories. The AI model automatically generates and publishes content according to the content publishing strategies corresponding to the accounts in the account matrix system; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning. 2. The method according to claim 1, characterized in that, When generating content for the sub-accounts using a large AI model, the following is included: Collect hotspot data; The hot data is understood by using a large AI model for decision-making to determine the appropriate content type, as well as the appropriate country/region, target audience, industry and/or product category, and to identify at least one target sub-account related to the target country/region, target audience, industry and/or product category; The AI model, pre-trained for the target sub-account, generates content of the corresponding type for the target sub-account and determines the publication time according to the corresponding publication time strategy. 3. The method according to claim 2, characterized in that, The target entities include cross-border commodity information service platforms; The content types include product recommendation categories; The method further includes: The target products related to the hot data and the relevant display materials of the target products are determined from the product information service platform. The relevant display materials for the target product are provided to the AI model trained for the target sub-account, so as to generate text content related to product recommendations based on the hot data, and assemble it with the display materials into product recommendation content to be published. 4. The method according to claim 2, characterized in that, The process of generating content for the sub-account using a large AI model also includes: After generating content of the corresponding type for the target sub-account, the generated content is adapted to the SNS network corresponding to the target sub-account in terms of content style, format, specifications, required interactive elements and/or user preferences. 5. The method according to claim 4, characterized in that, During the adaptation process, if the SNS network supports multiple different content formats, the generated content will be rewritten in multiple ways to generate content corresponding to multiple content formats, and then published in different content formats respectively. 6. The method according to claim 1, characterized in that it further comprises: The content publishing strategy for the sub-accounts will be updated based on user feedback regarding the content posted on the accounts. 7. The method according to claim 1, characterized in that it further comprises: The AI-powered large-scale model enables collaborative interaction among different accounts within the account matrix system. 8. The method according to claim 7, characterized in that, The collaborative interaction processing between different accounts in the account matrix system includes: The quality of the content generated by the AI model for the sub-account is judged. If it meets the preset conditions, the generated content mentions the country/region account of the sub-account and/or the main account, so as to realize the interaction between the sub-account and the country/region account and/or the main account. 9. The method according to claim 7, characterized in that, The collaborative interaction processing between different accounts in the account matrix system includes: The sub-accounts are used to forward some of the content posted by the country/region accounts of their respective countries/regions. 10. The method according to claim 7, characterized in that, The collaborative interaction processing between different accounts in the account matrix system includes: Sub-accounts with the same or related account profiles follow each other and/or forward each other's content. 11. The method according to claim 1, characterized in that it further comprises: After receiving comments or private messages from users regarding published content, accounts in the account matrix system use AI big data models to generate responses to the comments or private messages. 12. The method according to claim 1, characterized in that it further comprises: The system uses sub-accounts within the account matrix to follow well-known accounts outside the corresponding industry or field, and after these well-known accounts publish content, it uses an AI big data model to generate comments on the published content. 13. The method according to claim 1, characterized in that it further comprises: By setting up keyword and/or topic monitoring mechanisms, we can capture and analyze trending topics and/or discussion trends in SNS networks in real time, identify sub-accounts in the account matrix system that are related to the trending topics and/or discussion trends, and generate comments on the trending topics and/or discussion trends by the sub-accounts through an AI big data model. 14. A method for processing SNS account information, characterized in that it includes: The target interface receives account configuration information submitted by the target entity. The account configuration information includes the main account registered by the target entity in the SNS network, multiple accounts registered in multiple different countries/regions, and at least one sub-account associated with each country/region account. The sub-accounts include accounts registered by the target entity in the SNS network, and/or accounts registered by other entities in the SNS network and authorized to the target entity for operation and management. The sub-accounts have their own account profile positioning in terms of targeting specific sub-segments of people, industries and/or product categories. Based on the account configuration information, an SNS account matrix system for the target entity is constructed so that content can be automatically generated and published by an AI big data model according to the content publishing strategies corresponding to the accounts in the account matrix system; wherein, the content publishing strategy corresponding to the sub-account is related to the account profile positioning. 15. The method according to claim 14, characterized in that, The AI model is also used for collaborative interaction between different accounts. 16. A computer-readable storage medium having a computer program stored thereon, characterized in that, when executed by a processor, the program implements the steps of the method according to any one of claims 1 to 15. 17. An electronic device, characterized in that it comprises: One or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 15. 18. A computer program product comprising a computer program/computer executable instructions, characterized in that, when the computer program/computer executable instructions are executed by a processor in an electronic device, they implement the steps of the method according to any one of claims 1 to 15.