Content delivery method and device of social platform, storage medium and equipment

Abstract

The invention discloses a content delivery method and device of a social platform, a storage medium and equipment, and belongs to the technical field of deep learning. Acquiring a node to be scheduled in the strategy map, wherein the node is generated according to the delivery target and is associated with a plurality of delivery strategies; according to the platform data of the social platform, generating a platform ecological snapshot and a target crowd coverage degree; generating an account adaptation degree according to the release record of the user account; according to the platform ecological snapshot, the target population coverage degree and the account adaptation degree, screening the most adaptive n triads, wherein each triad comprises a delivery strategy, a social platform and a user account; and performing content delivery by using a corresponding user account in a corresponding social platform by adopting a delivery strategy corresponding to the triple. According to the method, the optimal'strategy-platform-account 'combination can be dynamically screened out, the content triggering precision and the interactive conversion efficiency are improved, the scheduling weight corresponding to the delivery effect is updated to the node, the node is scheduled according to the scheduling weight, and autonomous optimization of delivery is realized.

Description

Technical Field

[0001] This application relates to the field of deep learning technology, and in particular to a method, apparatus, storage medium and device for content delivery on a social platform. Background Technology

[0002] Currently, social media platforms generally use a combination of "content templates + scheduled posting + account pool polling" to achieve cross-platform content distribution. The basic process includes: (1) The operators configure the content template, target social platform (such as Weibo, Douyin, Facebook, etc.) and release time in the management backend, and add the generated scheduling task to the scheduling queue; (2) The task scheduler triggers the scheduling task based on the release time and selects the corresponding interface adaptation module according to the target social platform; (3) The interface adaptation module converts the content into a platform-supported format based on the content template (such as converting text and images into Weibo format and short videos into Douyin format). (4) The interface adaptation module quickly calls the target social platform's application programming interface (API) to complete the content delivery.

[0003] However, the above content delivery methods rely on static rules, such as fixed posting time, fixed account, and fixed content template, and lack the ability to adapt to differences in the platform ecosystem (such as content format, review mechanism, and user activity patterns), which results in the inability to achieve precise content delivery. Summary of the Invention

[0004] This application provides a content delivery method, apparatus, storage medium, and device for social media platforms, addressing the problem that content delivery using static rules lacks adaptability to platform ecosystem differences, resulting in inaccurate targeting. The technical solution is as follows: According to a first aspect of this application, a method for content delivery on a social media platform is provided, the method comprising: Obtain the nodes to be scheduled in the strategy graph. The nodes are generated according to the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format. Based on the platform data of each social platform to be targeted, a platform ecosystem snapshot and target audience coverage are generated. The platform ecosystem snapshot represents a quantitative snapshot of the social platform across multiple targeting dimensions, and the target audience coverage represents the overlap between the users of the social platform and the target audience of the content. An account fit score is generated based on the posting history of each user account to be deployed, whereby the account fit score represents the degree of fit between the user account and the content. Based on the platform ecosystem snapshot, the target audience coverage, and the account fit, select the n triples with the highest fit. Each triple includes a campaign strategy, a social platform, and a user account, where n is a positive integer. For each triple, the corresponding delivery strategy is adopted, and the content is delivered using the corresponding user account on the corresponding social platform.

[0005] In one possible implementation, the step of selecting the n triples with the highest fit based on the platform ecosystem snapshot, the target audience coverage, and the account fit includes: By combining all advertising strategies, social media platforms, and user accounts, multiple triples can be obtained; For each triplet, a platform fit score is generated based on the delivery strategy within the triplet and the platform ecosystem snapshot of the social platform. The platform fit score, target audience coverage, and account fit score corresponding to the triplet are weighted and calculated to obtain the triplet matching score. Select the n triplets with the highest scores from all triplet matching scores.

[0006] In one possible implementation, generating a platform ecosystem snapshot based on platform data from each social platform to be deployed includes: For each social media platform, periodically obtain the platform's user reach, user activity data, trending topics data, and sensitive data subject to review. The platform ecosystem snapshot is generated based on the population coverage, user activity data, trending data, and sensitive data.

[0007] In one possible implementation, generating account suitability based on the posting records of each user account to be deployed includes: For each user account to be targeted, obtain the user account's follower data, violation data, and posting records within the historical time period; The account fit is generated based on the fan data, the violation data, and the content quality and interaction volume of each post.

[0008] In one possible implementation, the step of employing a delivery strategy corresponding to each triple and delivering content using the corresponding user account on the corresponding social platform includes: For each triple, obtain the delivery strategy, social platform, and user account within the triple; Identify the interface adapter corresponding to the social platform; Using the interface adapter, the content to be delivered is converted into multimedia content that matches the publishing format in the delivery strategy; When the publication time specified in the delivery strategy is reached, the user account is used to publish the multimedia content on the social platform.

[0009] In one possible implementation, obtaining the node to be scheduled in the policy graph includes: Obtain the scheduling weight of each node in the policy graph; The nodes to be scheduled are selected based on the scheduling weights.

[0010] In one possible implementation, the method further includes: The delivery effect of the content is collected within a predetermined period after the content is delivered; The scheduling weight of the node is adjusted based on the delivery effect, and the scheduling weight is used to represent the scheduling value of the node.

[0011] According to a second aspect of this application, a content delivery device for a social platform is provided, the device comprising: The acquisition module is used to acquire nodes to be scheduled in the strategy graph. The nodes are generated according to the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format. The generation module is used to generate platform ecosystem snapshots and target audience coverage based on the platform data of each social platform to be targeted. The platform ecosystem snapshot represents a quantitative snapshot of the social platform in multiple targeting dimensions, and the target audience coverage represents the overlap between the users of the social platform and the target audience of the content. The generation module is also used to generate account fit scores based on the publishing records of each user account to be deployed, wherein the account fit score represents the fit between the user account and the content. The filtering module is used to filter the n triples with the highest fit based on the platform ecosystem snapshot, the target audience coverage and the account fit. Each triple includes a campaign strategy, a social platform and a user account, where n is a positive integer. The delivery module is used to deliver content to each triple using the corresponding delivery strategy on the corresponding social platform and with the corresponding user account.

[0012] According to a third aspect of this application, a computer-readable storage medium is provided, wherein at least one instruction is stored therein, the at least one instruction being loaded and executed by a processor to implement the content delivery method for a social platform as described above.

[0013] According to a fourth aspect of this application, a computer device is provided, the computer device including the content delivery device of the aforementioned social platform.

[0014] The beneficial effects of the technical solution provided in this application include at least the following: By constructing a ternary matching model that integrates platform ecosystem snapshots, target audience coverage, and account suitability, the optimal "strategy-platform-account" combination can be dynamically selected. This breaks through the limitations of traditional reliance on static rules or single-dimensional matching, enabling content to be delivered to the target audience on the most suitable social platform, by the most matching user account, and in the most appropriate publishing format and time, thereby significantly improving content triggering accuracy and interaction conversion efficiency.

[0015] The platform ecosystem snapshot quantifies the dynamic characteristics of each social platform in real time and combines the coverage of the target audience for multi-dimensional evaluation. This allows for the perception and adaptation to the ecological differences and real-time changes of different social platforms, enabling instant optimization of the advertising strategy and avoiding advertising failures caused by dynamic factors such as social platform algorithm adjustments and user behavior migration.

[0016] By dynamically generating account fit based on historical posting records and fine-tuning it in conjunction with specific campaign strategies, it can accurately identify and prioritize the allocation of high-value, highly matched account resources to execute high-potential content. This avoids high-quality accounts being occupied by inefficient strategies or low-quality accounts affecting campaign performance, thereby maximizing resource utilization efficiency and campaign revenue while protecting account health.

[0017] Based on the strategy graph, the scheduling weights generated based on the campaign performance are continuously updated to the nodes, forming a data-driven closed loop of "campaign-feedback-optimization". This allows high-performance strategies to receive higher priority in subsequent scheduling, driving the campaign strategies to continuously evolve towards higher efficiency, thereby achieving continuous self-learning and autonomous optimization in the campaign process. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0019] Figure 1 This is a flowchart of a content delivery method for a social platform provided in one embodiment of this application; Figure 2 This is a flowchart of a content delivery method for a social platform provided in one embodiment of this application; Figure 3This is a simplified flowchart illustrating a content delivery method for a social platform provided in one embodiment of this application; Figure 4 This is a structural block diagram of a content delivery device for a social platform provided in one embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0021] like Figure 1 The diagram illustrates a flowchart of a content delivery method for a social platform according to an embodiment of this application. This content delivery method for a social platform can be applied to a computer device. The content delivery method for the social platform may include: Step 101: Obtain the node to be scheduled in the strategy graph. This node is generated based on the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format.

[0022] The content to be promoted is content that needs to be promoted on social media platforms, and can cover multiple areas, including but not limited to products, brand information, educational information, community activities, official information, data reports, entertainment information, recruitment information, etc.

[0023] The targeting objective describes the specific information of the content to be targeted and the expected goals. This objective can include the target audience profile (e.g., "women aged 25-35 in first-tier cities"), content theme (e.g., "winter lipsticks that brighten skin"), expected interaction type (e.g., likes, comments, shares, favorites), and scheduling weight. The scheduling weight represents the scheduling value of a node, and there is a positive correlation between scheduling weight and scheduling value. That is, the higher the scheduling weight, the higher the scheduling value of that node, and the easier it is to target the content corresponding to that node; conversely, the lower the scheduling weight, the lower the scheduling value of that node, and the less likely it is to target the content corresponding to that node.

[0024] The campaign strategy includes the timing and format of the posting. The timing of the posting is when the content to be posted on the social media platform. The format of the posting is a multimedia format that matches the social media platform, which can include text, images, graphic text, long videos, short videos, etc.

[0025] Each node can be associated with multiple campaign strategies, each with different release times and / or formats. For example, one campaign strategy might be "8 AM text and image" while another is "9 PM short video".

[0026] When you need to deliver content on a social media platform, you can obtain the pre-configured delivery target for that content, generate a node in the strategy graph based on the delivery target, and finally associate multiple delivery strategies with that node.

[0027] In one example, the target audience profile is "women aged 18-30 who are interested in beauty and makeup and are often active at night". The content theme is "winter lipsticks that make your skin look brighter". The expected interaction type is "comment + favorite". The scheduling weight is 1. After generating a node for this target audience, it is associated with 3 targeting strategies: S1 is "text and image + 10 am", S2 is "short video + 8 pm", and S3 is "live broadcast announcement + 9 pm".

[0028] In this embodiment, at least one node that meets the scheduling conditions can be selected from all nodes in the strategy graph, and steps 102-105 are executed to deliver the content corresponding to these nodes. The scheduling conditions can be set according to actual business needs and are not limited here.

[0029] Step 102: Generate platform ecosystem snapshots and target audience coverage based on the platform data of each social platform to be targeted. The platform ecosystem snapshot represents a quantitative snapshot of the social platform across multiple targeting dimensions, while the target audience coverage represents the overlap between the social platform's users and the target audience for the content.

[0030] Social media platforms can include Weibo, TikTok, Facebook, etc.

[0031] Platform ecosystem snapshots depict a quantitative snapshot of key dimensions such as the content dissemination environment, user activity patterns, and platform rule status on a specific social platform at a particular moment.

[0032] Target audience coverage measures the degree of overlap between the user base of a social media platform and the target audience profile of the content to be delivered. The higher the value, the greater the proportion of the target audience that can be reached by delivering content on that social media platform, and the higher the potential reach efficiency.

[0033] The algorithms for platform ecosystem snapshots and target audience coverage are detailed in the following text and will not be elaborated here.

[0034] Step 103: Generate account fit scores based on the posting records of each user account to be deployed. Account fit scores represent the fit between the user account and the content.

[0035] Account fit is used to measure the potential effectiveness and overall suitability of a specific user account in performing campaign tasks. It is not a static, general score, but a dynamic, task-specific capability and risk assessment score.

[0036] The algorithm for account suitability is described in detail below and will not be repeated here.

[0037] Step 104: Select the n triples with the highest fit based on the platform ecosystem snapshot, target audience coverage, and account fit. Each triple includes a campaign strategy, a social platform, and a user account, where n is a positive integer.

[0038] In this embodiment, a triple can be formed by the targeting strategy, the social platform, and the user account, and the order of the targeting strategy, the social platform, and the user account in the triple is not limited. For example, the triple can be (targeting strategy, social platform, user account), (targeting strategy, user account, social platform), or (social platform, user account, targeting strategy), etc.

[0039] Taking the triplet (advertising strategy, social media platform, user account) as an example, the advertising strategy includes strategy 1 and strategy 2, the social media platform includes Weibo and Douyin, and the user account includes account 1 and account 2. Then the following triplets can be formed: (strategy 1, Weibo, account 1), (strategy 1, Weibo, account 2), (strategy 1, Douyin, account 1), (strategy 1, Douyin, account 2), (strategy 2, Weibo, account 1), (strategy 2, Weibo, account 2), (strategy 2, Douyin, account 1), (strategy 2, Douyin, account 2).

[0040] Specifically, the corresponding triplet matching score can be calculated based on the platform ecosystem snapshot, target audience coverage, and account suitability for each triplet. Then, the n triplet matching scores with the highest scores are selected from all the triplet matching scores, and the n triplets corresponding to the n triplet matching scores are taken as the n triplets with the highest matching degree.

[0041] Step 105: For each triplet, adopt the corresponding delivery strategy and use the corresponding user account to deliver content on the corresponding social platform.

[0042] For each of the n triples, read the delivery strategy, social platform, and user account within the triple, adopt the delivery strategy, and use the user account to deliver content on the social platform.

[0043] In summary, the content delivery method for social platforms provided in this application constructs a ternary matching model that integrates platform ecosystem snapshots, target audience coverage, and account suitability. This model can dynamically select the optimal "strategy-platform-account" combination, breaking through the limitations of traditional methods that rely on static rules or single-dimensional matching. This allows content to be delivered to the target audience on the most suitable social platform, by the most matching user account, and in the most appropriate publishing format and time, thereby significantly improving content triggering accuracy and interaction conversion efficiency.

[0044] The platform ecosystem snapshot quantifies the dynamic characteristics of each social platform in real time and combines the coverage of the target audience for multi-dimensional evaluation. This allows for the perception and adaptation to the ecological differences and real-time changes of different social platforms, enabling instant optimization of the advertising strategy and avoiding advertising failures caused by dynamic factors such as social platform algorithm adjustments and user behavior migration.

[0045] By dynamically generating account fit based on historical posting records and fine-tuning it in conjunction with specific campaign strategies, it can accurately identify and prioritize the allocation of high-value, highly matched account resources to execute high-potential content. This avoids high-quality accounts being occupied by inefficient strategies or low-quality accounts affecting campaign performance, thereby maximizing resource utilization efficiency and campaign revenue while protecting account health.

[0046] like Figure 2 The diagram illustrates a flowchart of a content delivery method for a social platform according to an embodiment of this application. This content delivery method for a social platform can be applied to a computer device. The content delivery method for the social platform may include: Step 201: Obtain the node to be scheduled in the strategy graph. The node is generated based on the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format.

[0047] The details regarding the content to be delivered, the delivery objectives, the delivery strategy, and the nodes are described above and will not be repeated here.

[0048] Specifically, obtaining nodes to be scheduled in the strategy graph can include: obtaining the scheduling weight of each node in the strategy graph; and filtering nodes to be scheduled based on the scheduling weight.

[0049] Each node in the strategy graph corresponds to a scheduling weight. Scheduling conditions can be set based on the scheduling weight, and then the scheduling conditions can be used to filter the nodes to be scheduled.

[0050] In one example, the scheduling condition can be set to a scheduling weight greater than a weight threshold. When filtering nodes to be scheduled, the scheduling weight of each node in the policy graph can be compared with this weight threshold. If the scheduling weight of a node is greater than the weight threshold, then the node is determined to be scheduled; if the scheduling weight of a node is less than or equal to the weight threshold, then the node is ignored. The weight threshold can be set according to actual business needs and is not limited here.

[0051] Step 202: Generate platform ecosystem snapshots and target audience coverage based on the platform data of each social platform to be targeted. The platform ecosystem snapshot represents a quantitative snapshot of the social platform across multiple targeting dimensions, while the target audience coverage represents the overlap between the social platform's users and the target audience for the content.

[0052] The explanations of the platform ecosystem snapshot and target audience coverage are detailed in the descriptions above and will not be repeated here.

[0053] Specifically, a platform ecosystem snapshot is generated based on the platform data of each social platform to be deployed. This may include: periodically acquiring the social platform's audience coverage, user activity data, trending data, and sensitive data for each social platform; and generating a platform ecosystem snapshot based on the audience coverage, user activity data, trending data, and sensitive data for each social platform.

[0054] Platform data includes audience coverage, user activity data, trending topics data, and data sensitive to content moderation. Audience coverage can be the proportion of people using the social media platform; user activity data can include active user accounts and active time periods; trending topics data can include popular topics, popular data, popular videos, popular events, popular products, etc.; and data sensitive to content moderation can include a database of sensitive keywords for content moderation and platform-specific rules.

[0055] Specifically, tools such as large models can be used to process audience coverage, user activity data, trending data, and sensitive data to obtain a snapshot of the platform ecosystem.

[0056] Target audience coverage measures the degree of overlap between the user group of a social platform and the target audience profile of the content to be delivered. It is a value between 0 and 1. The higher the value, the greater the proportion of the target audience that can be reached by delivering content on that social platform, and the higher the potential reach efficiency.

[0057] Step 203: Generate account fit score based on the posting records of each user account to be deployed. Account fit score represents the fit between the user account and the content.

[0058] Each social media platform corresponds to an account pool, and the account fit can be calculated for each available user account in that pool. The account fit is a value between 0 and 1.

[0059] Specifically, account fit can be generated based on the posting history of each user account to be targeted. This can include: for each user account to be targeted, obtaining the user account's follower data, violation data, and posting history within a historical period; and generating account fit based on the follower data, violation data, and the content quality and interaction volume of each posting.

[0060] Fan data can include the number of fans and fan level; violation data can include the number of times a user account has posted inappropriate content and the severity level; posting records can include the frequency of posted content, content quality, and number of interactions.

[0061] Specifically, tools such as large models can be used to process fan data, violation data, and posting records to obtain account fit.

[0062] In one example, the account pool contains five available user accounts: Account A1 (500,000 Weibo followers, high engagement rate in the evening), Account A2 (Douyin influencer, beauty category), Account A3 (overseas Facebook account, not targeting the intended audience), Account A4 (new Weibo account, no historical data), and Account A5 (Douyin secondary account, recent violations). Based on each user account's recent follower data, violation data, and 200 posting records, the account fit scores for accounts A1-A5 are calculated as follows: Account A1 = 0.95, Account A2 = 0.90, Account A3 = 0.40, Account A4 = 0.60, and Account A5 = 0.20.

[0063] By scoring user accounts, high-scoring accounts A1 and A2 are assigned to high-match strategies, while accounts A4 and A5 are not selected. This prevents high-value user accounts from being used by low-potential strategies, protects account health, and improves return on investment (ROI).

[0064] Step 204: Combine all campaign strategies, social media platforms, and user accounts to obtain multiple triples.

[0065] Each triple consists of a campaign strategy, a social media platform, and a user account. For example, if the campaign strategies include Strategy 1 and Strategy 2, the social media platforms include Weibo and Douyin, and the user accounts include Account 1 and Account 2, then the following triples can be formed: (Strategy 1, Weibo, Account 1), (Strategy 1, Weibo, Account 2), (Strategy 1, Douyin, Account 1), (Strategy 1, Douyin, Account 2), (Strategy 2, Weibo, Account 1), (Strategy 2, Weibo, Account 2), (Strategy 2, Douyin, Account 1), (Strategy 2, Douyin, Account 2).

[0066] Step 205: For each triplet, generate platform fit based on the ad placement strategy within the triplet and the platform ecosystem snapshot of the social platform. Calculate the platform fit, target audience coverage, and account fit corresponding to the triplet by weighting the calculation to obtain the triplet matching score.

[0067] In this embodiment, tools such as large models can be used to process the platform ecosystem snapshot to obtain the platform adaptability score. The platform adaptability score is a value between 0 and 1. In one example, Weibo user activity peaks between 8-10 PM, so the platform adaptability score P is... weibo =0.92; Douyin beauty short videos have high dissemination efficiency between 7-9 pm, so the platform compatibility P is higher. douyin =0.89; Facebook has low coverage among the target audience, so the platform fit is P. fb =0.3.

[0068] After obtaining platform compatibility, account compatibility, and target audience coverage, you can iterate through all the "strategy-social platform-user account" triples and calculate the triple matching score.

[0069] Specifically, the triplet matching score S = α·P + β·U + γ·A, where P is the platform adaptability, U is the target audience coverage, A is the account adaptability, and α, β, and γ are adjustable coefficients that can be set according to business needs; no specific value is set in this embodiment. In one example, α = 0.4, β = 0.4, and γ = 0.2.

[0070] Step 206: Select the n triplets with the highest scores from all triplet matching scores, where n is a positive integer.

[0071] In this embodiment, the n triplet matching scores with the highest scores can be selected from all triplet matching scores. The n triplets corresponding to the n triplet matching scores are taken as the n triplets with the highest matching degree, and the corresponding strategy instantiation is triggered.

[0072] In one example, if n=2, then (strategy S2, Douyin, account A2) and (strategy S1, Weibo, account A1) can be selected as the two triples with the highest matching degree.

[0073] Step 207: For each triplet, adopt the corresponding delivery strategy and use the corresponding user account to deliver content on the corresponding social platform.

[0074] Specifically, for each triple, the corresponding delivery strategy is adopted, and the content is delivered using the corresponding user account on the corresponding social platform. This may include: for each triple, obtaining the delivery strategy, social platform, and user account within the triple; determining the interface adapter corresponding to the social platform; using the interface adapter to convert the content to be delivered into multimedia content that matches the publishing format in the delivery strategy; and when the publishing time in the delivery strategy is reached, publishing the multimedia content on the social platform using the user account.

[0075] In practice, strategies and content can be converted into the platform's native format based on the type of social platform (such as generating long articles for Weibo, subtitled videos for Douyin, or Facebook Carousel for the same content), and then the content can be published on user accounts using the social platform's API.

[0076] For example, a 15-second lipstick swatch video can be generated based on strategy S2 and released at 8 pm on December 1st to coincide with the platform's peak traffic and improve exposure efficiency; a text and image blog post with product images can be generated based on strategy S1 and released at 10 am on December 1st.

[0077] Step 208: Collect the content delivery effect within a predetermined time period after the content is delivered; adjust the scheduling weight of the node based on the delivery effect, whereby the scheduling weight is used to represent the scheduling value of the node.

[0078] The scheduled time period can be set according to business needs, such as 24 hours, but this embodiment does not limit it.

[0079] Campaign performance can be measured by statistics on the expected types of interaction, such as impressions, clicks, conversions, and favorites.

[0080] The performance of the campaign is positively correlated with the scheduling weight; that is, the better the performance, the higher the scheduling weight, and the worse the performance, the lower the scheduling weight.

[0081] In one example, if 1200 comments and favorites are collected on Douyin videos and 400 interactions are collected on Weibo within 24 hours, the scheduling weight of the node corresponding to strategy S2 can be increased to 1.2, while the scheduling weight of the node corresponding to strategy S1 can be maintained at 1.0. This ensures that in the next scheduling cycle, the node with the higher scheduling weight will receive a higher scheduling priority, forming a positive feedback loop.

[0082] By updating the scheduling weights of nodes in the policy graph, high-performing policies are given higher priority in subsequent scheduling, thus forming a data-driven policy evolution.

[0083] like Figure 3 As shown, the system generates an ecosystem awareness (platform ecosystem snapshot) from platform behavior data, a dynamic strategy graph from operational configurations (targeting objectives), and account scores (account suitability and target audience coverage) from account resources. A triplet matching engine then calculates the matching score for each triplet, selecting the n triplets with the highest scores. Finally, a heterogeneous adapter converts the n triplets into a format compatible with the corresponding social media platform for publication. After content publication, feedback can be collected, and the scheduling weights of nodes in the strategy graph can be updated based on this feedback to ensure that nodes with higher scheduling weights receive higher scheduling priority.

[0084] In summary, the content delivery method for social platforms provided in this application constructs a ternary matching model that integrates platform ecosystem snapshots, target audience coverage, and account suitability. This model can dynamically select the optimal "strategy-platform-account" combination, breaking through the limitations of traditional methods that rely on static rules or single-dimensional matching. This allows content to be delivered to the target audience on the most suitable social platform, by the most matching user account, and in the most appropriate publishing format and time, thereby significantly improving content triggering accuracy and interaction conversion efficiency.

[0085] The platform ecosystem snapshot quantifies the dynamic characteristics of each social platform in real time and combines the coverage of the target audience for multi-dimensional evaluation. This allows for the perception and adaptation to the ecological differences and real-time changes of different social platforms, enabling instant optimization of the advertising strategy and avoiding advertising failures caused by dynamic factors such as social platform algorithm adjustments and user behavior migration.

[0086] By dynamically generating account fit based on historical posting records and fine-tuning it in conjunction with specific campaign strategies, it can accurately identify and prioritize the allocation of high-value, highly matched account resources to execute high-potential content. This avoids high-quality accounts being occupied by inefficient strategies or low-quality accounts affecting campaign performance, thereby maximizing resource utilization efficiency and campaign revenue while protecting account health.

[0087] Based on the strategy graph, the scheduling weights generated based on the campaign performance are continuously updated to the nodes, forming a data-driven closed loop of "campaign-feedback-optimization". This allows high-performance strategies to receive higher priority in subsequent scheduling, driving the campaign strategies to continuously evolve towards higher efficiency, thereby achieving continuous self-learning and autonomous optimization in the campaign process.

[0088] like Figure 4 The diagram illustrates a structural block diagram of a content delivery device for a social platform according to an embodiment of this application. This content delivery device can be applied to a computer device and includes: The acquisition module 410 is used to acquire nodes to be scheduled in the strategy graph. The nodes are generated according to the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format. The generation module 420 is used to generate a platform ecosystem snapshot and target audience coverage based on the platform data of each social platform to be targeted. The platform ecosystem snapshot represents a quantitative snapshot of the social platform in multiple targeting dimensions, and the target audience coverage represents the overlap between the users of the social platform and the target audience of the content. The generation module 420 is also used to generate account fit scores based on the publishing records of each user account to be deployed, wherein the account fit score represents the fit between the user account and the content. The filtering module 430 is used to filter the n triples with the highest fit based on the platform ecosystem snapshot, the target audience coverage and the account fit. Each triple includes a campaign strategy, a social platform and a user account, where n is a positive integer. The delivery module 440 is used to deliver content to each triple using the delivery strategy corresponding to the triple and the corresponding user account on the corresponding social platform.

[0089] In an optional embodiment, the filtering module 430 is further configured to: By combining all advertising strategies, social media platforms, and user accounts, multiple triples can be obtained; For each triplet, a platform fit score is generated based on the delivery strategy within the triplet and the platform ecosystem snapshot of the social platform. The platform fit score, target audience coverage, and account fit score corresponding to the triplet are weighted and calculated to obtain the triplet matching score. Select the n triplets with the highest scores from all triplet matching scores.

[0090] In an optional embodiment, the generation module 420 is further configured to: For each social media platform, periodically obtain the platform's user reach, user activity data, trending topics data, and sensitive data subject to review. The platform ecosystem snapshot is generated based on the population coverage, user activity data, trending data, and sensitive data.

[0091] In an optional embodiment, the generation module 420 is further configured to: For each user account to be targeted, obtain the user account's follower data, violation data, and posting records within the historical time period; The account fit is generated based on the fan data, the violation data, and the content quality and interaction volume of each post.

[0092] In an optional embodiment, the delivery module 440 is further configured to: For each triple, obtain the delivery strategy, social platform, and user account within the triple; Identify the interface adapter corresponding to the social platform; Using the interface adapter, the content to be delivered is converted into multimedia content that matches the publishing format in the delivery strategy; When the publication time specified in the delivery strategy is reached, the user account is used to publish the multimedia content on the social platform.

[0093] In an optional embodiment, the acquisition module 410 is further configured to: Obtain the scheduling weight of each node in the policy graph; The nodes to be scheduled are selected based on the scheduling weights.

[0094] In an optional embodiment, the device further includes: The data collection module is used to collect the delivery effect of the content within a predetermined time period after the content is delivered; An adjustment module is used to adjust the scheduling weight of the node based on the delivery effect, wherein the scheduling weight is used to represent the scheduling value of the node.

[0095] In summary, the content delivery device for social platforms provided in this application can dynamically select the optimal "strategy-platform-account" combination by constructing a three-element matching model that integrates platform ecosystem snapshots, target audience coverage, and account suitability. This breaks through the limitations of traditional methods that rely on static rules or single-dimensional matching, enabling content to be delivered to the target audience on the most suitable social platform, by the most matching user account, and in the most appropriate publishing format and time, thereby significantly improving the accuracy of content triggering and the efficiency of interaction conversion.

[0096] The platform ecosystem snapshot quantifies the dynamic characteristics of each social platform in real time and combines the coverage of the target audience for multi-dimensional evaluation. This allows for the perception and adaptation to the ecological differences and real-time changes of different social platforms, enabling instant optimization of the advertising strategy and avoiding advertising failures caused by dynamic factors such as social platform algorithm adjustments and user behavior migration.

[0097] By dynamically generating account fit based on historical posting records and fine-tuning it in conjunction with specific campaign strategies, it can accurately identify and prioritize the allocation of high-value, highly matched account resources to execute high-potential content. This avoids high-quality accounts being occupied by inefficient strategies or low-quality accounts affecting campaign performance, thereby maximizing resource utilization efficiency and campaign revenue while protecting account health.

[0098] Based on the strategy graph, the scheduling weights generated based on the campaign performance are continuously updated to the nodes, forming a data-driven closed loop of "campaign-feedback-optimization". This allows high-performance strategies to receive higher priority in subsequent scheduling, driving the campaign strategies to continuously evolve towards higher efficiency, thereby achieving continuous self-learning and autonomous optimization in the campaign process.

[0099] One embodiment of this application provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the content delivery method for a social platform as described above.

[0100] One embodiment of this application provides a computer device, which includes a content delivery device for any of the aforementioned social media platforms.

[0101] It should be noted that the content delivery device for social platforms provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the content delivery device for social platforms can be divided into different functional modules to complete all or part of the functions described above. In addition, the content delivery device for social platforms provided in the above embodiments and the content delivery method embodiments for social platforms belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0102] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0103] The above description is not intended to limit the embodiments of this application. Any adjustments, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A method for content delivery on a social media platform, characterized in that, The method includes: Obtain the nodes to be scheduled in the strategy graph. The nodes are generated according to the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format. Based on the platform data of each social platform to be targeted, a platform ecosystem snapshot and target audience coverage are generated. The platform ecosystem snapshot represents a quantitative snapshot of the social platform across multiple targeting dimensions, and the target audience coverage represents the overlap between the users of the social platform and the target audience of the content. An account fit score is generated based on the posting history of each user account to be deployed, whereby the account fit score represents the degree of fit between the user account and the content. Based on the platform ecosystem snapshot, the target audience coverage, and the account fit, select the n triples with the highest fit. Each triple includes a campaign strategy, a social platform, and a user account, where n is a positive integer. For each triple, the corresponding delivery strategy is adopted, and the content is delivered using the corresponding user account on the corresponding social platform.

2. The content delivery method for a social platform according to claim 1, characterized in that, The process of selecting the n triples with the highest fit based on the platform ecosystem snapshot, the target audience coverage, and the account fit includes: By combining all advertising strategies, social media platforms, and user accounts, multiple triples can be obtained; For each triplet, a platform fit score is generated based on the delivery strategy within the triplet and the platform ecosystem snapshot of the social platform. The platform fit score, target audience coverage, and account fit score corresponding to the triplet are weighted and calculated to obtain the triplet matching score. Select the n triplets with the highest scores from all triplet matching scores.

3. The content delivery method for a social platform according to claim 1, characterized in that, The process of generating a platform ecosystem snapshot based on platform data from each social media platform to be deployed includes: For each social media platform, periodically obtain the platform's user reach, user activity data, trending topics data, and sensitive data subject to review. The platform ecosystem snapshot is generated based on the population coverage, user activity data, trending data, and sensitive data.

4. The content delivery method for a social platform according to claim 1, characterized in that, The process of generating account suitability based on the posting records of each user account to be deployed includes: For each user account to be targeted, obtain the user account's follower data, violation data, and posting records within the historical time period; The account fit is generated based on the fan data, the violation data, and the content quality and interaction volume of each post.

5. The content delivery method for a social platform according to claim 1, characterized in that, For each triple, the corresponding delivery strategy is adopted, and content is delivered using the corresponding user account on the corresponding social platform, including: For each triple, obtain the delivery strategy, social platform, and user account within the triple; Identify the interface adapter corresponding to the social platform; Using the interface adapter, the content to be delivered is converted into multimedia content that matches the publishing format in the delivery strategy; When the publication time specified in the delivery strategy is reached, the user account is used to publish the multimedia content on the social platform.

6. The content delivery method for a social platform according to claim 1, characterized in that, The step of obtaining the node to be scheduled in the strategy graph includes: Obtain the scheduling weight of each node in the policy graph; The nodes to be scheduled are selected based on the scheduling weights.

7. The content delivery method for a social platform according to any one of claims 1 to 6, characterized in that, The method further includes: The delivery effect of the content is collected within a predetermined period after the content is delivered; The scheduling weight of the node is adjusted based on the delivery effect, and the scheduling weight is used to represent the scheduling value of the node.

8. A content delivery device for a social platform, characterized in that, The device includes: The acquisition module is used to acquire nodes to be scheduled in the strategy graph. The nodes are generated according to the delivery target of the content to be delivered, and each node is associated with multiple delivery strategies, including the delivery time and delivery format. The generation module is used to generate platform ecosystem snapshots and target audience coverage based on the platform data of each social platform to be targeted. The platform ecosystem snapshot represents a quantitative snapshot of the social platform in multiple targeting dimensions, and the target audience coverage represents the overlap between the users of the social platform and the target audience of the content. The generation module is also used to generate account fit scores based on the publishing records of each user account to be deployed, wherein the account fit score represents the fit between the user account and the content. The filtering module is used to filter the n triples with the highest fit based on the platform ecosystem snapshot, the target audience coverage and the account fit. Each triple includes a campaign strategy, a social platform and a user account, where n is a positive integer. The delivery module is used to deliver content to each triple using the corresponding delivery strategy on the corresponding social platform and with the corresponding user account.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, which is loaded and executed by a processor to implement the content delivery method for a social platform as described in any one of claims 1 to 7.

10. A computer device, characterized in that, The computer device includes: the content delivery device for the social platform as described in claim 8.

Images