Information delivery plan processing method and electronic device

Abstract

Embodiments of the present application disclose an information putting plan processing method and an electronic device. The method comprises: receiving putting strategy information submitted by a first user, and generating a parent plan related to information putting according to the putting strategy information, wherein the putting strategy information comprises a plurality of strategy elements selected by the first user and at least one execution item provided for the strategy elements; calling a first artificial intelligence (AI) module to analyze the execution items of the plurality of strategy elements, determining a strategy element that can be split into execution items and a corresponding splitting scheme, and generating a plurality of sub-plans according to the splitting scheme, and creating a plurality of corresponding sub-plans in a third-party information putting system, so that the third-party information putting system puts information based on the sub-plans. Through the embodiments of the present application, the richness of advertisement creation can be realized, the efficiency is improved, and the cost is reduced.

Description

Technical Field

[0001] This application relates to the field of information processing technology, and in particular to methods and electronic devices for processing information delivery plans. Background Technology

[0002] For merchants in product information service systems (or "e-commerce platforms"), advertising through external media (such as social networking platforms and content platforms) is a crucial path to achieving scalable growth in metrics like click-through rates and sales. Typically, there are multiple media outlets, each associated with its own advertising platform (referred to as a "media-side advertising platform"). Merchants can log into different media-side advertising platforms to publish their ads. Alternatively, to improve efficiency, some marketing service platforms (providing advertisers (merchants) with one-stop digital marketing solutions) offer cross-media advertising services. In this case, the marketing service platform can also be called an "aggregated advertising platform," which uniformly connects to the media-side advertising platforms of multiple different media outlets. Merchants can use this aggregated advertising platform to create advertising plans. Then, the aggregated platform can copy the specific advertising plan information to the corresponding media-side advertising platforms by calling their interfaces, and the media platforms then execute the advertising tasks.

[0003] While the aforementioned aggregated advertising platforms can prevent merchants from individually connecting with multiple media outlets, the creation of specific advertising plans still relies entirely on manual work by the merchant or agency team. To achieve better advertising results, the same product may require multiple different advertising plans to be created and launched separately. In this case, the merchant or agency team needs to manually create multiple advertising plans, resulting in relatively low efficiency. Summary of the Invention

[0004] This application provides a method for processing information delivery plans and electronic equipment, which can enrich the creation of advertisements, improve efficiency, and reduce costs.

[0005] This application provides the following solution: A method for processing information delivery plans, the method being applied to an aggregated delivery system, comprising: Receive the delivery strategy information submitted by the first user, and generate a parent plan related to the delivery of information based on the delivery strategy information. The delivery strategy information includes multiple strategy elements selected by the first user, and at least one execution item provided for the strategy elements. The first artificial intelligence (AI) module is invoked to perform execution item analysis on the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans according to the splitting schemes. Then, multiple corresponding sub-plans are created in the third-party information delivery system so that the third-party information delivery system can deliver information based on the sub-plans.

[0006] The strategy element that can be split into execution items can be one or more, and the execution items under the same strategy element can be split into multiple groups. Multiple sub-plans can be generated by cross-combining the execution item grouping results between different strategy elements.

[0007] The creation of multiple corresponding sub-plans in the third-party information delivery system includes: Based on the aforementioned splitting scheme, a pre-packaged workflow is invoked to complete the steps of creating multiple corresponding sub-plans in a third-party information delivery system. The workflow is generated by pre-arranging and encapsulating multiple atomic tools according to a preset logical flow.

[0008] This includes the process of creating multiple corresponding sub-plans in a third-party information delivery system, which involves calling the interfaces provided by the third-party information delivery system. The method further includes: If an interface call exception occurs, the second AI module is triggered to perform exception type identification and call the corresponding repair tool to repair the exception.

[0009] This also includes: Upon detecting an abnormal interface call, the execution of the first AI module is interrupted; After the second AI module completes the anomaly repair, it wakes up the first AI module to continue the sub-plan creation process.

[0010] This also includes: Upon receiving a notification message from a third-party information delivery system indicating that a page access request has been received, the corresponding second user information is determined based on the device identifier associated with the page access request. Based on the second user information, a preset number of sub-plans are selected from the multiple sub-plans, and after preliminary sorting, they are provided to the third-party information delivery system so that the third-party information delivery system can determine the target sub-plan from the preset number of sub-plans and display it in the target resource position on the current page.

[0011] This also includes: When providing the pre-selected number of sub-plans to the third-party information delivery system, the system predicts the first probability of the materials associated with the sub-plan from being clicked to being converted into the next action point under the current page access request conditions, and provides this prediction to the third-party information delivery system. The third-party information delivery system is also used to predict the second probability of the materials associated with the sub-plan from being exposed to being clicked under the current page access request conditions, so that after making a joint bid based on the first probability and the second probability, the sub-plans are fine-tuned to determine the sub-plans that are actually exposed for the current page access request.

[0012] This also includes: The system obtains real-time feedback records of the actual exposure and click data of the sub-plan from the third-party information delivery system. If the click-through rate of the sub-plan is lower than the threshold, the sub-plan is modified or deleted.

[0013] 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.

[0014] 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 any of the preceding methods.

[0015] 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.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed: Through the embodiments of this application, an AI module (Agent) can be provided in the aggregated delivery system. A first user can submit delivery strategy information in the aggregated delivery system, which may include multiple strategy elements selected by the first user and at least one execution item provided for each strategy element. Subsequently, the aggregated delivery system can generate a parent plan related to information delivery based on the delivery strategy information, and call the first AI module to analyze the execution items of the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans based on the splitting schemes. These sub-plans are then created for a third-party information delivery system, enabling the third-party information delivery system to deliver information based on the sub-plans. In this way, the richness of ad creation can be achieved by splitting the parent plan into multiple different sub-plans. Furthermore, since the first user does not need to create each sub-plan, efficiency can be improved and costs reduced.

[0017] In the implementation process, to enhance intelligence, atomic tools can be provided to the AI ​​module. Furthermore, for high-frequency, stable operations, they can be encapsulated into workflows. For example, the process of creating a sub-plan for a third-party information delivery system can be encapsulated as a workflow. This eliminates the need for the AI ​​module to monitor the execution of multiple different atomic tools within a specific workflow; it can invoke the workflow as if it were a single, ordinary atomic tool, thereby further improving efficiency. This approach strikes a balance between atomic tools and encapsulated workflows, ensuring business determinism while granting the AI ​​module flexible invocation capabilities.

[0018] Furthermore, if an interface call anomaly is detected during the execution of the first AI module, the second AI module can be invoked to identify the specific anomaly type and call the corresponding repair tool to fix it. Through this AI-based anomaly classifier and repair strategy library, the system can automatically repair common problems within seconds, improving the completion rate of project creation. Moreover, the aforementioned first and second AI modules enable automation, intelligence, and self-healing of the project creation process.

[0019] Furthermore, it can enable joint bidding between the aggregated advertising system and third-party information advertising systems. The third-party information advertising system can also provide real-time feedback on the actual exposure and click data of sub-campaigns. The aggregated advertising system can perform statistics, and if the click-through rate of a sub-campaign is lower than the threshold, it can trigger the modification or deletion of the sub-campaign, realizing the replacement and optimization of sub-campaigns based on real-time advertising performance, and achieving a closed loop of "perception-decision-execution-reflection".

[0020] 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

[0021] 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.

[0022] Figure 1 This is a schematic diagram of the system architecture provided in the embodiments of this application; Figure 2 This is a flowchart of the method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the interaction process provided in the embodiments of this application; Figure 4 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0023] 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.

[0024] As described in the background section, in the prior art, after a merchant formulates an advertising plan through an aggregated advertising system, the aggregated advertising system can call the interface of a media-side advertising system (in this embodiment, it can be uniformly referred to as a "third-party information advertising system," with the merchant and the aggregated advertising platform being the first and second parties, respectively) to copy a mirror image of the advertising plan one-to-one in the third-party information advertising system, thereby placing the advertisement on the third-party information advertising system. If the same product (or other promotional target) needs to be promoted through multiple advertising plans, the merchant needs to formulate multiple advertising plans in the aggregated advertising system, and then the aggregated advertising system will copy these advertising plans to the third-party information advertising system respectively. However, the process of manually creating advertising plans by the merchant consumes a significant amount of time and effort.

[0025] To address the above issues, this application provides an AI-powered intelligent infrastructure system based on an Agent (artificial intelligence module, or, as some other term, intelligent agent) architecture. This system aims to address the challenges of creating rich content for off-site advertisements while simultaneously reducing costs and improving delivery efficiency. Specifically, in this application embodiment, as... Figure 1As shown, multiple AI modules can be provided on the aggregation and delivery system side. For example, this could include a first AI module for plan creation, a second AI module for integrity assurance, and so on. These AI modules can be called Agents or intelligent agents, etc. Such AI modules can be intelligent programs based on AI models (e.g., AI language models), capable of performing related intelligent processing, calling tools (including interfaces, knowledge bases, etc.), and realizing contextual interaction between different AI modules. In specific implementations, the execution of the second AI module is typically triggered only when an interface call exception occurs during the creation process.

[0026] Based on the aforementioned AI module, merchants and other first-time users can submit specific campaign strategy information within a dedicated ad aggregation system. This system can then create a parent campaign based on this information. The specific campaign strategy information can include multiple strategy elements selected by the first-time user, and at least one action item provided for each strategy element. These strategy elements, often referred to as strategy components, are typically defined by a third-party ad delivery system. The ad aggregation system can then replicate some or all of these components, which become the available options for the first-time user when creating an ad campaign through the system. For example, specific strategy elements or components might include "express targeting," "express bid," "express budget," "select ad placement" (the same media provider may offer various placement options such as dynamic feed pages and content aggregation pages), "express creative" (users can upload images, videos, etc., to express creative ideas), or features like "check it out" or "browse" to add action points to creative materials. In this application embodiment, the first user can provide multiple different execution items for the same strategy element. For example, when "expressing targeting", multiple different groups of people can be selected; when "expressing creativity", multiple different materials can be uploaded, and the material types can also be multiple, such as both pictures and videos; when "selecting a placement", multiple different placements such as dynamic information flow pages and content aggregation pages can be selected, and so on.

[0027] After receiving the aforementioned campaign strategy information submitted by the first user, the aggregated campaign system can generate a parent campaign accordingly. Additionally, it can invoke the aforementioned first AI module. This first AI module can analyze multiple strategy elements based on knowledge related to campaign information, determining strategy elements that can be split into execution items and their corresponding splitting schemes. The specific knowledge information can exist in the form of an external knowledge base for the first AI module, or it can be provided to the first AI module through other means. When analyzing the aforementioned knowledge information, the first AI module can determine whether certain key strategy elements include multiple different execution items. If so, it may identify these key strategy elements as those that can be split into execution items. For example, under the strategy element "Expressing Creativity," there are various types of materials, including both image and video materials, and there may be multiple materials of the same type. In this case, the "Expressing Creativity" element can be a sub-item that can be split into actionable items. Specific splitting schemes can include: firstly, splitting by material type, with different types of materials being assigned to different sub-plans; secondly, limiting the number of materials of the same type, for example, no more than three of the same type of material in the same sub-plan. If there are more than three of the same type of material, these materials can be combined in pairs, or in groups of three, resulting in multiple material combinations, and so on. These different material combinations can all be assigned to different sub-plans.

[0028] Besides creative elements, elements related to "selecting placement" are also generally suitable for splitting execution items. For example, if a campaign strategy has multiple execution items under the "selecting placement" element, including dynamic information feed pages and content aggregation pages, then different placements can be divided into different sub-campaigns, and so on. Furthermore, other strategy elements can also be split according to different execution items in principle. For example, if the "expression targeting" element includes multiple audiences, it can also be split into execution items, and so on, depending on the specific needs.

[0029] After determining which strategy elements can be split according to different execution items, and how to split each strategy element, the splitting results of different strategy elements can be cross-combined to form multiple different sub-plans. For example, suppose a campaign strategy includes 4 strategy elements, two of which have been split into execution items: element 1's execution items are split into 4 groups, and element 2's execution items are split into 2 groups. Then, the 4 groups of execution items under element 1 and the 2 groups of execution items under element 2 can be cross-combined, and together with other unsplit elements, to form complete campaign strategy information. Subsequent campaigns can be generated based on this splitting and recombination.

[0030] After splitting the campaign into multiple sub-plans on the aggregated delivery system side, corresponding sub-plans can be created on the third-party information delivery system. In other words, in this embodiment, the first user creates a parent plan in the aggregated delivery system, but the aggregated delivery system can split it into multiple sub-plans and then copy them to the specific third-party information delivery system.

[0031] In practical implementation, fine-grained atomicity tools can be provided to handle the splitting and creation of child plans from parent plans. Additionally, high-frequency, stable operations can be encapsulated into workflows. That is, a workflow can include multiple atomicity tools with predefined execution order, input data construction methods, etc. Therefore, the first AI module can also invoke such workflows to complete processes such as creating corresponding child plans for third-party information delivery systems.

[0032] In addition, in this embodiment of the application, anomaly detection can also be performed during the interaction with the third-party information delivery system. If an anomaly occurs, the aforementioned second AI module can be automatically triggered. The second AI module can identify the specific anomaly type and call the corresponding repair tool to repair the anomaly, so as to ensure the integrity of the process.

[0033] Furthermore, when a PV (Page View) request is generated on the media side, the aggregated advertising system can filter out multiple sub-campaigns suitable for advertising on that page, perform preliminary sorting, and then provide them to the third-party information delivery system. The third-party information delivery system can then select one sub-campaign according to bidding rules and display its corresponding creative materials in the dedicated ad placement on the specific page. Afterwards, the third-party information delivery system can provide real-time feedback on the exposure and click data of the specific sub-campaigns to the aggregated advertising system. Based on the performance of the specific sub-campaigns, the aggregated advertising system can adjust, delete, or replace them, eliminating inefficient sub-campaigns. Additionally, when bidding on specific sub-campaigns, a joint bidding approach can be used between the aggregated advertising system and the third-party information delivery system.

[0034] The specific implementation schemes provided in the embodiments of this application will be described in detail below.

[0035] First, this application provides an information delivery plan processing method. This method is applied to an aggregated delivery system, which can be located between a first user and a third-party information delivery system, helping the first user complete information delivery to the external third-party information delivery system. For details, see... Figure 2 The method may include: S201: Receive the delivery strategy information submitted by the first user, and generate a parent plan related to the delivery of information based on the delivery strategy information, wherein the delivery strategy information includes multiple strategy elements selected by the first user, and at least one execution item provided for the strategy elements.

[0036] In this embodiment, because an aggregated delivery system exists, the first user only needs to interact with this system and does not need to directly interact with the third-party information delivery system. The aggregated delivery system will then interface with the specific third-party information delivery system. Specifically, if the first user needs to deliver information, they can submit delivery strategy information to the aggregated delivery system. The aggregated delivery system can provide the first user with a specific information delivery interface, which may include multiple optional strategy elements (components). The first user can choose to use these strategy elements and select, input, or upload specific execution items under each strategy element to complete the submission of the delivery strategy information. For example, the first user can use the aggregated delivery system's interface to express targeting, budget, creative ideas (submitting specific materials, etc.), select ad placements, etc. In this embodiment, the first user can configure multiple different execution items under the same strategy element, including selecting multiple ad placements, submitting multiple different materials, etc.

[0037] After receiving the delivery strategy information submitted by the first user, the aggregated delivery system can first generate a parent plan based on this information. Here, the parent plan is mainly in contrast to subsequent sub-plans, referring to a plan directly created based on the information submitted by the first user. However, in this embodiment, instead of directly copying this parent plan to the third-party delivery system, the system first performs a plan splitting process, dividing it into multiple sub-plans, and then copies these sub-plans to the third-party delivery system.

[0038] S202: The first artificial intelligence (AI) module is invoked to perform execution item analysis on the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans according to the splitting schemes. Then, multiple corresponding sub-plans are created in the third-party information delivery system so that the third-party information delivery system can deliver information based on the sub-plans.

[0039] After generating the parent plan, the first AI module can be invoked to perform plan splitting. Specifically, knowledge information related to information delivery can be provided to the first AI module through external knowledge bases, including situations in which execution items can be split, etc. After being invoked, the first AI module can determine the strategy elements for splitting execution items and the corresponding splitting schemes based on the aforementioned knowledge information.

[0040] The strategy elements that can be split into execution items can be one or more. Execution items under the same strategy element can be split into multiple groups. The splitting scheme is generated by combining the grouping results of execution items with different strategy elements. For example, suppose the element "expressing creativity" corresponds to multiple execution items, including multiple images and videos, etc. In this case, splitting can be done first based on the type of material. If there are many materials of the same type, they can be divided into multiple groups according to the quantity, etc., so that different types of materials can be divided into different sub-plans. Each sub-plan includes only materials of the same type, and the quantity can not exceed a certain value, etc. In addition, execution items can also be split for page positions. For example, in the above example, execution items for different page positions such as dynamic information flow pages and content aggregation pages can be divided into different sub-plans, etc. After determining the elements that can be split into execution items and the splitting method, the grouping results of execution items can be cross-combined with different strategy elements to generate multiple sub-plans.

[0041] After splitting the campaign into multiple sub-campaigns on the aggregated delivery system side, corresponding sub-campaigns can be created on a third-party information delivery system. In other words, in this embodiment, the first user creates a parent campaign in the aggregated delivery system; however, the aggregated delivery system can split it into multiple sub-campaigns and then copy them to the specific third-party information delivery system. This approach allows for greater richness in ad creation through multiple different sub-campaigns. Furthermore, since the first user does not need to create each sub-campaign individually, efficiency is improved and costs are reduced.

[0042] From another perspective, in this application embodiment, when the first user submits specific delivery strategy information, he / she can provide multiple different execution items under the same strategy element without considering whether these execution items are suitable for coexistence in the same plan. Subsequently, the first AI module in the aggregated delivery system will split the specific execution items. In addition, during the splitting process, specific knowledge information will be referenced to split execution items that are not suitable for coexistence in the same plan into different sub-plans.

[0043] In practical implementation, to better automate processes such as splitting specific deployment plans, fine-grained atomic tools can also be provided, such as... Figure 3 The specific atomic tools mentioned above can include creating accounts, initializing accounts, modifying plans, querying plans, updating budgets, and correcting parameters, etc. These atomic tools can be used to perform the process of splitting and creating sub-plans from parent plans. Of course, a single operation may require the cooperation of multiple atomic tools. For example, after determining the splitting scheme and obtaining multiple sub-plans, the process of creating multiple corresponding sub-plans for a third-party information delivery system typically requires the use of plan parsing (parsing the sub-plans on the aggregation delivery system side), field mapping (mapping fields in the aggregation delivery system to the field formats and value ranges required for interface calls in the third-party information delivery system), media identity verification, calling the third-party information delivery system interface, result recording and status synchronization (establishing a mapping relationship between the sub-plan IDs on the aggregation delivery system side and the corresponding sub-plan IDs created by the third-party information delivery system based on the sub-plan IDs returned by the third-party information delivery system), etc. If only the above atomic tools are defined, the first AI module, after generating the splitting scheme, needs to orchestrate these atomic tools to determine their order, input data format, etc., before calling the atomic tools.

[0044] However, considering that the processes of creating sub-plans for third-party information delivery systems are frequent occurrences in this embodiment, and that the atomic tools and their order used in the specific processing are generally relatively stable, these high-frequency, stable operations can be encapsulated as workflows. That is, a workflow can include multiple atomic tools with predefined execution order, input data construction methods, etc. Therefore, after the first AI module determines the specific splitting scheme and generates multiple sub-plans on the aggregation delivery system side, this workflow can be invoked to complete the process of creating corresponding sub-plans for third-party information delivery systems. In this way, the first AI module no longer needs to focus on the execution status of multiple different atomic tools within the specific workflow; it can invoke the workflow as if it were a single, ordinary atomic tool, thereby further improving efficiency.

[0045] Besides the sub-plan creation process mentioned above, which can be encapsulated as a workflow, other operations characterized by high frequency and stability can also be encapsulated as workflows, such as creating unit / creative workflows or editing unit workflows, etc. Of course, for other less frequent or stable combinations, they can continue to exist as atomic tools. In this way, a balance can be achieved between atomic tools and encapsulated workflows, ensuring business determinism while giving the AI ​​module flexible invocation capabilities.

[0046] During the process of creating multiple sub-plans for the third-party information delivery system, anomalies may occur due to interface call failures in the third-party information delivery system. Under existing technology, users may be unaware of such anomalies arising during the interaction between the aggregated delivery system and the third-party information delivery system, making it impossible to confirm whether the plan was successfully created in the third-party information delivery system. However, in this embodiment, anomaly detection can be performed during the interaction with the third-party information delivery system. If an anomaly occurs, the aforementioned second AI module can be automatically triggered. This second AI module identifies the specific anomaly type and calls the corresponding repair tool to fix the anomaly (such as retrying interface calls that failed due to network jitter). Through this AI-based anomaly classifier and repair strategy library, the system can automatically repair common problems within seconds, improving the completeness of plan creation.

[0047] Furthermore, in the event of an detected interface call anomaly, the execution of the first AI module can be interrupted first, and then the second AI module can be triggered to perform specific anomaly type identification and repair. After the repair is completed, the first AI module can be automatically triggered to continue the subsequent process. During this process, operation order preservation can be achieved (the system's state changes and execution order remain logically consistent before and after the first AI module is interrupted). Combined with the interrupt recovery mechanism, data consistency can be ensured in high-concurrency scenarios, and the ability to resume long-running tasks from breakpoints can be supported, improving system robustness. In summary, through the aforementioned first and second AI modules, the automation, intelligence, and self-healing of the plan creation process can be achieved.

[0048] It should be noted that exceptions caused by reasons such as failed API calls do not necessarily need to be communicated to the first user; that is, the first user does not need to be aware of the occurrence of such exceptions. Of course, for other types of exceptions that prevent the process from continuing, the first user can be notified. For example, during the creation of a sub-plan with a third-party information delivery system, if the system finds that the materials in the sub-plan are identical or highly similar to those in other already created plans, potentially causing confusion, the first user can be notified to resubmit the materials, and so on.

[0049] After creating a sub-plan for the third-party information delivery system, when a PV (Page View) request is generated on the media side (e.g., a user on the media side visits a page in the media client), the third-party information delivery system can provide the device identifier of the request initiator to the aggregated delivery system. The aggregated delivery system identifies the specific second user based on the device identifier, determines multiple sub-plans suitable for delivery to that user based on this second user information, performs a preliminary ranking, and then provides them to the third-party information delivery system. For example, it can determine whether the second user's demographics, tags, etc., align with the targeting objectives of the specific sub-plan. In addition, the specific sub-plan may also correspond to other execution items such as the delivery time period. Therefore, it is also possible to combine execution item information under multiple factors to filter out multiple sub-plans that meet the conditions, perform a preliminary ranking, and then provide them to the third-party information delivery system.

[0050] Third-party information delivery systems can select a sub-campaign based on bidding rules and display its corresponding materials in a dedicated ad placement on a specific page. The third-party system can then provide real-time feedback on the sub-campaign's impressions and clicks to the aggregated delivery system. Based on the sub-campaign's performance, the aggregated delivery system can adjust, delete, or replace it, eliminating inefficient sub-campaigns.

[0051] In addition, when bidding on specific sub-campaigns, a joint bidding approach can be used between the aggregated advertising system and the third-party information delivery system. For example, the third-party information delivery system can predict the click-through rate (CTR) of the specific sub-campaign's creative materials under the current page conditions, while the aggregated advertising system can predict the probability of the specific sub-campaign being converted from a click into a further action point such as a purchase under the current page conditions, i.e., the conversion rate. Then, a joint bid can be made on the specific sub-campaign based on the above CTR and conversion rate. The third-party information delivery system can then refine the ranking of multiple sub-campaigns based on the joint bid results to determine which specific sub-campaign to expose on the current page.

[0052] In other words, when the aggregated delivery system provides a preset number of sub-campaigns to the third-party information delivery system, it can predict the first probability of the sub-campaign's associated creatives from being clicked to being converted into the next action point (including purchase, etc.) under the current page access request conditions, and provide this to the third-party information delivery system. The third-party information delivery system is also used to predict the second probability of the sub-campaign's associated creatives from being exposed to being clicked under the current page access request conditions, so that after making a joint bid based on the first probability and the second probability, the sub-campaigns are fine-tuned to determine the sub-campaigns that will actually be exposed to the current page access request.

[0053] In addition, third-party information delivery systems can provide real-time feedback on the actual exposure and click data of sub-campaigns, aggregated delivery systems can perform statistical analysis, and the first AI module can observe the execution results and reflect on them. If the click-through rate of a sub-campaign falls below a threshold, it can trigger modifications or deletion of the sub-campaign. Specifically, modifications to sub-campaigns can include changing creatives, adjusting targeting, and so on. This can be achieved by calling corresponding atomic tools.

[0054] In summary, through the embodiments of this application, an AI module (Agent) can be provided in the aggregated delivery system. A first user can submit delivery strategy information in the aggregated delivery system, which may include multiple strategy elements selected by the first user and at least one execution item provided for each strategy element. Subsequently, the aggregated delivery system can generate a parent plan related to information delivery based on the delivery strategy information, and call the first AI module to analyze the execution items of the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans based on the splitting schemes. These sub-plans are then created for a third-party information delivery system, enabling the third-party information delivery system to deliver information based on the sub-plans. In this way, the richness of ad creation can be achieved by splitting the parent plan into multiple different sub-plans. Furthermore, since the first user does not need to create each sub-plan, efficiency can be improved and costs reduced.

[0055] In the implementation process, to enhance intelligence, atomic tools can be provided to the AI ​​module. Furthermore, for high-frequency, stable operations, they can be encapsulated into workflows. For example, the process of creating a sub-plan for a third-party information delivery system can be encapsulated as a workflow. This eliminates the need for the AI ​​module to monitor the execution of multiple different atomic tools within a specific workflow; it can invoke the workflow as if it were a single, ordinary atomic tool, thereby further improving efficiency. This approach strikes a balance between atomic tools and encapsulated workflows, ensuring business determinism while granting the AI ​​module flexible invocation capabilities.

[0056] Furthermore, if an interface call anomaly is detected during the execution of the first AI module, the second AI module can be invoked to identify the specific anomaly type and call the corresponding repair tool to fix it. Through this AI-based anomaly classifier and repair strategy library, the system can automatically repair common problems within seconds, improving the completion rate of project creation. Moreover, the aforementioned first and second AI modules enable automation, intelligence, and self-healing of the project creation process.

[0057] Furthermore, it can enable joint bidding between the aggregated advertising system and third-party information advertising systems. The third-party information advertising system can also provide real-time feedback on the actual exposure and click data of sub-campaigns. The aggregated advertising system can perform statistics, and if the click-through rate of a sub-campaign is lower than the threshold, it can trigger the modification or deletion of the sub-campaign, realizing the replacement and optimization of sub-campaigns based on real-time advertising performance, and achieving a closed loop of "perception-decision-execution-reflection".

[0058] 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.).

[0059] Corresponding to the foregoing method embodiments, this application also provides an information delivery plan processing apparatus, which is applied to an aggregated delivery system and includes: The parent plan creation unit is used to receive the delivery strategy information submitted by the first user and generate a parent plan related to the delivery of information based on the delivery strategy information. The delivery strategy information includes multiple strategy elements selected by the first user and at least one execution item provided for the strategy elements. The plan splitting unit is used to call the first artificial intelligence (AI) module to perform execution item analysis on the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans according to the splitting schemes. Then, it creates multiple corresponding sub-plans in the third-party information delivery system so that the third-party information delivery system can deliver information based on the sub-plans.

[0060] The strategy element that can be split into execution items can be one or more, and the execution items under the same strategy element can be split into multiple groups. Multiple sub-plans can be generated by cross-combining the execution item grouping results between different strategy elements.

[0061] In practical implementation, the plan splitting unit can be used specifically for: Based on the aforementioned splitting scheme, a pre-packaged workflow is invoked to complete the steps of creating multiple corresponding sub-plans in a third-party information delivery system. The workflow is generated by pre-arranging and encapsulating multiple atomic tools according to a preset logical flow.

[0062] This includes the process of creating multiple corresponding sub-plans in a third-party information delivery system, which involves calling the interfaces provided by the third-party information delivery system. The device may further include: The anomaly repair unit is used to sense the interface call situation. If an interface call anomaly occurs, it triggers the second AI module to perform anomaly type identification and calls the repair tool corresponding to the anomaly type to perform anomaly repair.

[0063] Additionally, it may include: An interrupt execution unit is used to interrupt the execution of the first AI module after detecting an interface call exception; The wake-up unit is used to wake up the first AI module to continue the sub-plan creation process after the second AI module completes the anomaly repair.

[0064] Additionally, the device may also include: The second user information determination unit is used to determine the corresponding second user information based on the device identifier associated with the page access request when receiving a notification message from a third-party information delivery system that a page access request has been received. The sub-plan recall unit is used to filter out a preset number of sub-plans from the plurality of sub-plans based on the second user information, and provide them to the third-party information delivery system after preliminary sorting, so that the third-party information delivery system can determine the target sub-plan from the preset number of sub-plans and display it in the target resource position on the current page.

[0065] Furthermore, it may also include: The joint bidding unit is used to predict, when providing the selected preset number of sub-plans to the third-party information delivery system, the first probability of the creative associated with the sub-plan from being clicked to being converted into the next action point under the current page access request conditions, and provide it to the third-party information delivery system. The third-party information delivery system is also used to predict the second probability of the creative associated with the sub-plan from being exposed to being clicked under the current page access request conditions, so that after joint bidding based on the first probability and the second probability, the sub-plans are fine-tuned and the actual sub-plans exposed for the current page access request are determined.

[0066] The sub-plan management unit is used to obtain the actual exposure and click records of the sub-plan in real time from the third-party information delivery system. If the click-through rate of the sub-plan is lower than the threshold, the sub-plan will be modified or deleted.

[0067] 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.

[0068] 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.

[0069] 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.

[0070] in, Figure 4 An exemplary architecture of an electronic device is shown, which may include a processor 410, a video display adapter 411, a disk drive 412, an input / output interface 413, a network interface 414, and a memory 420. The processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420 can communicate with each other via a communication bus 430.

[0071] The processor 410 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in this application.

[0072] The memory 420 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 420 can store the operating system 421 for controlling the operation of the electronic device 400, and the basic input / output system (BIOS) for controlling the low-level operations of the electronic device 400. Additionally, it can store a web browser 423, a data storage management system 424, and an information delivery plan processing system 425, etc. The aforementioned information delivery plan processing system 425 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 420 and executed by the processor 410.

[0073] Input / output interface 413 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.

[0074] Network interface 414 is used to connect a communication module (not shown in the figure) to enable communication and interaction 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.).

[0075] Bus 430 includes a pathway for transmitting information between various components of the device, such as processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420.

[0076] It should be noted that although the above-described device only shows the processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, memory 420, bus 430, 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.

[0077] 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.

[0078] 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.

[0079] The information delivery plan processing method and electronic device provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only 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

1. A method for processing information delivery plans, characterized in that, The method is applied to an aggregated delivery system, including: Receive the delivery strategy information submitted by the first user, and generate a parent plan related to the delivery of information based on the delivery strategy information. The delivery strategy information includes multiple strategy elements selected by the first user, and at least one execution item provided for the strategy elements. The first artificial intelligence (AI) module is invoked to perform execution item analysis on the multiple strategy elements, determine the strategy elements that can be split into execution items and the corresponding splitting schemes, and generate multiple sub-plans according to the splitting schemes. Then, multiple corresponding sub-plans are created in the third-party information delivery system so that the third-party information delivery system can deliver information based on the sub-plans.

2. The method according to claim 1, characterized in that, The strategy element that can be split into execution items can be one or more, and execution items under the same strategy element can be split into multiple groups. Multiple sub-plans can be generated by cross-combining the execution item grouping results between different strategy elements.

3. The method according to claim 1, characterized in that, The creation of multiple corresponding sub-plans in the third-party information delivery system includes: Based on the aforementioned splitting scheme, a pre-packaged workflow is invoked to complete the steps of creating multiple corresponding sub-plans in a third-party information delivery system. The workflow is generated by pre-arranging and encapsulating multiple atomic tools according to a preset logical flow.

4. The method according to claim 1, characterized in that, The process of creating multiple corresponding sub-plans in a third-party information delivery system includes calling the interfaces provided by the third-party information delivery system; The method further includes: If an interface call exception occurs, the second AI module is triggered to perform exception type identification and call the corresponding repair tool to repair the exception.

5. The method according to claim 4, characterized in that, Also includes: Upon detecting an abnormal interface call, the execution of the first AI module is interrupted; After the second AI module completes the anomaly repair, it wakes up the first AI module to continue the sub-plan creation process.

6. The method according to claim 1, characterized in that, Also includes: Upon receiving a notification message from a third-party information delivery system indicating that a page access request has been received, the corresponding second user information is determined based on the device identifier associated with the page access request. Based on the second user information, a preset number of sub-plans are selected from the multiple sub-plans, and after preliminary sorting, they are provided to the third-party information delivery system so that the third-party information delivery system can determine the target sub-plan from the preset number of sub-plans and display it in the target resource position on the current page.

7. The method according to claim 6, characterized in that, Also includes: When providing the pre-selected number of sub-plans to the third-party information delivery system, the system predicts the first probability of the materials associated with the sub-plan from being clicked to being converted into the next action point under the current page access request conditions, and provides this prediction to the third-party information delivery system. The third-party information delivery system is also used to predict the second probability of the materials associated with the sub-plan from being exposed to being clicked under the current page access request conditions, so that after making a joint bid based on the first probability and the second probability, the sub-plans are fine-tuned to determine the sub-plans that are actually exposed for the current page access request.

8. The method according to claim 6, characterized in that, Also includes: The system obtains real-time feedback records of the actual exposure and click data of the sub-plan from the third-party information delivery system. If the click-through rate of the sub-plan is lower than the threshold, the sub-plan is modified or deleted.

9. 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 described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: One or more processors; as well as 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 8.

11. 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 8.

Images