Information delivery plan processing method and system and electronic equipment
By receiving user dialogue content through the artificial intelligence (AI) module, understanding and generating optimization solutions, the problem of the intelligent advertising algorithm being in a "black box" state has been solved, enabling merchants to personalize and improve the stability of advertising.
Patent Information
- Application Number
- CN202510992550.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-17
- Publication Date
- 2025-12-09
AI Technical Summary
In existing advertising delivery systems, the intelligent delivery algorithm is in a "black box" state, making it difficult for merchants to directly intervene, resulting in unstable delivery results and difficulty in personalized optimization.
By receiving user conversations through the artificial intelligence (AI) module, understanding user intent, generating optimization plans, and intervening in the processing elements of the intelligent advertising algorithm, such as keywords, audiences, and bids, it provides an "end-to-end" solution.
It enables merchants to optimize the intelligent advertising algorithm and make personalized adjustments based on their needs, thereby improving the personalization and stability of the advertising results.
Smart Images
Figure CN121094884A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of information processing technology, and in particular to methods, systems and electronic devices for processing information delivery plans. Background Technology
[0002] Product promotion typically refers to advertising selected products to increase their exposure and give them more opportunities to be seen by consumers. For example, the decision to display the promoted product in a webpage ad slot can be based on whether the consumer's search keywords match the keywords configured for the promoted product, or whether the consumer's demographic matches the demographic configured for the promoted product, as well as the specific bidding situation of the promoted product for the corresponding keywords and demographics.
[0003] In traditional advertising methods, merchants can typically handle campaigns manually (including manually selecting keywords, target audiences, and setting bids), or they can choose a platform's intelligent campaign service. This means the platform uses its backend algorithm to select keywords and target audiences for the specific product to be promoted, and automatically sets bids. Of course, to prevent situations where manually selected keywords / target audiences fail to match user search terms / audiences, merchants can often combine this with services like "Intelligent Traffic Selection." This involves the backend algorithm using the merchant's manually selected keywords / target audiences as seeds, and then selecting more keywords / target audiences for further targeting, thus maximizing product exposure. Furthermore, besides selecting keywords, target audiences, and bids, the choice of advertising venues also plays a role.
[0004] In summary, whether it's "manual advertising + intelligent traffic selection" or "smart advertising," both involve algorithmic keyword / audience selection, algorithmic bidding, and algorithmic selection of advertising venues. While these smart advertising services bring convenience to merchants, they also create new problems. For example, in the process of providing smart advertising services through backend algorithms, there may be unstable consumption or spending, and the advertising performance may not meet the merchant's expectations. Faced with this situation, merchants often find it difficult to find suitable ways to resolve or improve the situation, and usually can only express their dissatisfaction through complaints.
[0005] To address the above issues, existing technologies include diagnostic tools designed to diagnose problems arising during ad campaigns and provide corresponding suggestions. However, these tools typically only diagnose known issues from a list of planned problems compiled offline by the query algorithm, and they rely on preset answer formats, directly obtaining pre-defined solutions through the tool's underlying query list. While some diagnostic tools can offer optimization suggestions, these suggestions usually only provide a few preset optional objectives, such as "new user acquisition" or "recovery of churned customers." Merchants can only choose specific optimization directions from these objectives, making it difficult to provide personalized optimization solutions tailored to each customer. Summary of the Invention
[0006] This application provides a method, system, and electronic equipment for processing information delivery plans, which can solve customers' marketing needs "end-to-end" and achieve "tunable intelligence".
[0007] This application provides the following solution:
[0008] A method for processing information delivery plans, comprising:
[0009] During the execution of the target information delivery plan, the system receives dialogue content submitted by users regarding the target information delivery plan; the target information delivery plan is executed partially or entirely through intelligent delivery algorithms.
[0010] The AI module is used to understand the dialogue content to determine the user's intent. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, then the reasonableness of the user's intent is judged.
[0011] If the rationality assessment is passed, the AI module determines the optimization scheme corresponding to the user's intent, and optimizes the target information delivery plan according to the optimization scheme.
[0012] The step of receiving the dialogue content submitted by the user regarding the target information delivery plan includes:
[0013] During the execution of the target information delivery plan, a dialogue entry point is provided for communicating with the AI module, so as to receive user-inputted dialogue content optimized for the target information delivery plan through the dialogue entry point.
[0014] This also includes:
[0015] Before the user inputs dialogue content, an inspection report on the target information delivery plan is provided. The inspection report includes data on the performance of the target information delivery plan on multiple inspection items, so that the user can express optimization needs based on the inspection results.
[0016] This also includes:
[0017] The dialogue entry point receives user input regarding the intelligent diagnosis of the target information delivery plan.
[0018] After understanding the received dialogue content and determining the user's intent through an AI model, the system analyzes the problems and corresponding reasons in the delivery process of the target information delivery plan, generates diagnostic results, and allows users to express their optimization needs based on the diagnostic results.
[0019] This also includes:
[0020] The AI model generates content to guide users in optimizing the target information delivery plan.
[0021] This also includes:
[0022] The AI model generates suggested directions for optimizing the target information delivery plan.
[0023] The intervention in the processing object elements of the intelligent delivery algorithm includes:
[0024] The intelligent targeting algorithm can be controlled to either block or emphasize specific keywords, audiences, or target areas when intelligently selecting keywords, audiences, or target areas.
[0025] The intervention in the processing object elements of the intelligent delivery algorithm includes:
[0026] The intelligent bidding algorithm adjusts the bid coefficients for specified keywords / audiences or time periods during intelligent bidding.
[0027] An information delivery planning and processing system, comprising:
[0028] The dialogue module is used to receive dialogue content submitted by users regarding the target information delivery plan during the execution of the target information delivery plan; the target information delivery plan is executed partially or entirely through intelligent delivery algorithms, and the dialogue content is used to express the needs generated during the execution of the target information delivery plan;
[0029] The AI traffic routing module is used to understand the dialogue content to determine the user's intent. If the user's intent is to diagnose the target information delivery plan, the traffic is routed to the AI diagnosis module. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, the user's intent is judged for reasonableness. If the reasonableness judgment is passed, the traffic is routed to the AI optimization module.
[0030] The AI diagnostic module is used to analyze the problems and corresponding causes that exist in the target information delivery plan during the delivery process, and generate diagnostic results.
[0031] The AI optimization module is used to determine the optimization scheme corresponding to the user intent, and optimize the target information delivery plan according to the optimization scheme.
[0032] The AI diagnostic module is also used to generate content to guide users in optimizing the target information delivery plan.
[0033] The AI triage module and the AI diagnosis module are the same module.
[0034] The AI optimization module includes multiple AI optimization sub-modules, each corresponding to a different processing object element;
[0035] The AI tuning module is also used to break down the user intent into multiple optimization tasks, each optimization task corresponding to a different processing object element, and to assign the optimization task to the corresponding AI tuning sub-module according to the processing object element information, so as to generate optimization schemes and perform optimization processing.
[0036] 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.
[0037] An electronic device, comprising:
[0038] One or more processors; and
[0039] A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the preceding descriptions.
[0040] 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.
[0041] According to the specific embodiments provided in this application, the following technical effects are disclosed:
[0042] Through the embodiments of this application, for information delivery plans that are partially or entirely executed through intelligent delivery algorithms, users can submit dialogue content for communication with the AI module during the plan execution process. Optimization can be achieved by intervening in the processing elements of the intelligent delivery algorithm. After receiving the user's dialogue content, if the AI module identifies the aforementioned intent, it can determine an optimization scheme corresponding to the user's intent after a reasonableness assessment, and optimize the target information delivery plan according to the optimization scheme to achieve an "end-to-end" solution to customer marketing needs. This solution makes the previously completely "black box" intelligent delivery algorithm more manageable, allowing users to express specific optimization requests in a "white box" manner. This addresses problems in the delivery process from a more fundamental perspective, enabling personalized optimization of information delivery plans based on the specific needs of merchants, rather than being limited to a few selectable options. Because optimization plans can be generated based on the "white-box" dialogue content of merchants and users, more merchant preference data can be expressed. Therefore, the intelligent delivery algorithm involved in the information delivery plan becomes a kind of "tunable intelligence".
[0043] 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
[0044] 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.
[0045] Figure 1 This is a schematic diagram of the system architecture provided in the embodiments of this application;
[0046] Figure 2 This is a flowchart of the method provided in the embodiments of this application;
[0047] Figure 3 This is a schematic diagram of the first interface provided in an embodiment of this application;
[0048] Figure 4 This is a schematic diagram of the second interface provided in an embodiment of this application;
[0049] Figure 5This is a schematic diagram of the system provided in the embodiments of this application;
[0050] Figure 6 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation
[0051] 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.
[0052] First, it should be noted that, in the process of developing this application, the inventors discovered that when information delivery plans (where information can specifically be advertisements, and the information delivery plan can refer to an advertising delivery plan, etc., are completed partially or entirely through intelligent delivery algorithms, the specific methods employed by the intelligent delivery algorithm in selecting keywords, selecting target audiences, and intelligently bidding significantly impact the actual advertising results. Moreover, the reasons for anomalies or discrepancies between advertising delivery and customer expectations may be partly due to user-defined settings (e.g., manually selected keywords or target audiences may be inaccurate); and partly due to the intelligent delivery algorithm itself. For example, when using intelligent delivery algorithms in part, the algorithm uses manually set keywords as seeds to generate more keywords and then delivers specific products based on these generated keywords. However, the generated keywords may be unreasonable, or some keywords may not be what the user wants. Alternatively, the intelligent bidding algorithm might set a lower bid coefficient for a certain time period during automatic bidding, but users might actually want to bid higher during that period to increase their product's exposure. In short, poor campaign performance caused by the backend intelligent bidding algorithm can manifest in the following ways:
[0053] a. Allocation of limited system resources: frequency adjustment, rough ranking scoring, frequency of adding and deleting words, etc.;
[0054] b. Mismatch between global parameters and customer's personalized needs: such as parameters for bid cap, time period preferences, optimization target preferences, etc.
[0055] c. Algorithm capabilities: reasonableness of bidding parameters, accuracy of scoring, real-time recall, and sudden changes in traffic in small and medium-sized promotion slots.
[0056] However, in existing technologies, whether using intelligent delivery algorithms partially or entirely, the specific intelligent delivery algorithms remain a "black box" for users. This means that users cannot intervene in these backend algorithms, and even if diagnostic tools identify some superficial problems, they cannot directly solve the problems at the underlying level.
[0057] To address the above issues, this application provides a solution for optimizing advertising campaigns using AI models (Artificial Intelligence, typically referring to deep learning models with massive parameters, which can store and process large amounts of information, thus achieving higher performance across various tasks). This allows users to express specific optimization needs through natural language dialogue, including optimizing the processing elements (also known as "targeting elements") within the intelligent advertising algorithm. In other words, this solution makes the backend algorithm operable, allowing users to express the various processing elements involved in the intelligent advertising algorithm in a "white-box" manner. These processing elements can include keywords, target audiences, bidding strategies, and targeting environments.
[0058] For example, suppose a merchant wants to promote a bicycle and sets keywords such as "mountain bike" and "cross-country bike." The algorithm might generate more keywords like "competitive bike" and "recreational bike." However, because "recreational bike" has a low relevance to the product, if the promoted product is displayed in the ad slot when a consumer searches for "recreational bike," the actual conversion rate of the user to purchase the product will not be good, making the resulting advertising cost a waste. In this case, users can interact with the AI model by inputting "please filter out words like 'recreational bike' when selecting keywords," and so on.
[0059] Besides specifying which keywords to block in the intelligent advertising algorithm, merchants can also interact with the AI model to input optimization methods such as blocking certain audiences or advertising scenarios. Alternatively, they can input optimization methods that focus more on specific types of keywords, audiences, or advertising scenarios. Furthermore, merchants can intervene in the bidding process of the intelligent advertising algorithm, including increasing or decreasing the bid coefficient for a particular time period, and so on.
[0060] Building upon the aforementioned approach that allows merchants to express their optimization needs regarding the processing elements of the underlying algorithm in a "white-box" manner, improvements can also be made to the specific ad delivery process. Specifically, after a user submits a specific optimization request, the AI model can identify the user's intent. If it determines that the user requires optimization of the processing elements within the underlying algorithm, the model can assess the reasonableness of the user's intent. If reasonable, the AI model can generate a specific optimization plan, such as which keywords or audiences to block or prioritize, and by how much to increase or decrease the bid coefficient. This optimization plan can then be directly applied to the ad delivery plan, achieving end-to-end optimization of the ad delivery plan.
[0061] Through the above methods, the processing elements of the intelligent advertising algorithm in the advertising campaign can be intervened, and users can express specific intervention methods in a "white-box" manner. This allows for the resolution of problems in the advertising process from a deeper level, enabling the advertising campaign to be optimized "personalized" according to the specific needs of each merchant user, rather than being limited to a few optional options. In other words, the solution provided in this application, because it can generate optimization plans based on the "white-box" dialogue content of merchants and users, allows more merchant preference data to be expressed. Therefore, the intelligent advertising algorithm involved in the advertising campaign becomes a "tunable intelligence."
[0062] Furthermore, since the specific optimization process requires merchants to express which elements(s) need optimization in what direction, some merchants may not know how to express this, or may not know whether their advertising campaign has room for optimization. Therefore, in the preferred approach, an inspection service can also be provided to users. That is, the performance data of certain inspection items in the user's advertising campaign can be statistically analyzed to determine if there are any problems. Merchants can then express their specific optimization needs based on this inspection report.
[0063] Alternatively, besides optimizing the processing elements in the intelligent advertising algorithm within the advertising campaign, this application embodiment can also provide AI model-based diagnostic services. That is, unlike the aforementioned inspection service, the diagnostic service provided in this application embodiment can be implemented based on an AI model. It can not only diagnose whether there are problems in certain projects, but also perform specific cause analysis, allowing users to obtain reference information and express their optimization needs more accurately. Furthermore, during the diagnostic process, it can also determine whether there are problems with the specific intelligent advertising algorithm. If so, it can recommend specific optimization directions, such as "spending rate optimization" and "cost optimization." Users can choose from these recommendations to submit specific optimization requests, or, based on the diagnostic results, express their preferred optimization direction by inputting specific dialogue content, and so on.
[0064] In practical implementation, since intelligent delivery algorithms may involve multiple processing object elements, optimizing different processing object elements may require different expertise in areas such as optimization scheme generation and execution. Furthermore, while diagnostic services can be provided as an option, they are relatively independent capabilities from tuning. Therefore, in practical applications, such as... Figure 1 As shown, this can be achieved using multiple AI modules (which can be called "intelligent agents," etc.). For example, it can first include an AI traffic routing module (also called a traffic routing agent), which identifies the user's intent based on the dialogue input, determining whether the user needs diagnosis or optimization. If diagnosis is needed, the request can be routed to the AI diagnosis module; otherwise, it can be routed to the AI optimization module (also called an optimization agent). Additionally, the AI traffic routing module can also perform a feasibility check on the optimization request. If the check passes, the request is then routed to the specific AI optimization module for processing. The AI optimization module can include multiple sub-agents, each corresponding to a specific processing object element, such as a keyword optimization sub-agent, an audience (targeting) optimization sub-agent, a bid optimization sub-agent, a site traffic optimization sub-agent, and so on. In this way, during the specific optimization process, the AI optimization module can also determine which type or type of processing object elements the user needs to optimize, and break it down into multiple optimization tasks. Each optimization task corresponds to a different processing object element. Based on the processing object element information, the optimization task is assigned to the corresponding AI optimization sub-module to generate optimization schemes and perform optimization processing.
[0065] The specific implementation schemes provided in the embodiments of this application will be described in detail below.
[0066] Example 1
[0067] First, this first embodiment provides a method for processing information delivery plans, see [link to example]. Figure 2 The method may specifically include:
[0068] S201: During the execution of the target information delivery plan, receive dialogue content submitted by the user regarding the target information delivery plan; the target information delivery plan is executed intelligently, partially or entirely, through an intelligent delivery algorithm.
[0069] The solution provided in this application embodiment allows for the diagnosis and optimization of an advertising campaign after a merchant has submitted or initiated a specific advertising campaign (during which specific products to be promoted have been set, and whether intelligent or manual + intelligent methods have been selected, etc.). The specific advertising campaign utilizes intelligent advertising algorithms at least partially, and the optimization primarily refers to optimizing the processing elements within the intelligent advertising algorithm, which was originally in a "black box" state, according to the merchant's "white-box" optimization requests.
[0070] In practical implementation, a dialogue entry point can be provided for users to interact with the AI module. Users can input dialogue content through this entry point, or they can first initiate a diagnostic request through the dialogue entry point, and after the diagnosis is completed, input specific dialogue content for optimization, and so on. The specific dialogue content can then be used to express the need to optimize the target advertising campaign.
[0071] There are several ways to provide a dialogue entry point. For example, one method is to provide a dialogue entry point for specific advertising campaigns within the list display interface of a merchant's advertising campaigns. In the campaign list interface, the specific dialogue entry point can exist in the form of a button element, etc. When the user hovers the cursor over the button element corresponding to the dialogue entry point, a pop-up or similar interface for inputting specific requests can be displayed.
[0072] Alternatively, the AI diagnostic and AI optimization services provided in this application embodiment can also be offered to users in a productized form, for example, the product name could be "**Escort". In this case, a specific dialogue entry can also be provided on the product's homepage or other interface, allowing users to first enter the product's homepage or other interface, and then input specific requests for diagnostics or optimization of a particular advertising campaign within the product.
[0073] The above describes a way for users to express themselves proactively. In other words, during the execution of a target advertising campaign, a dialogue entry point can be provided for communicating with the AI module. This dialogue entry point can be used to receive user-inputted dialogue content that is optimized for the target advertising campaign.
[0074] Besides user-initiated expression, since some users may not know how to express their goals or are unaware that their plans have room for optimization, a "marketing assistant" can also be provided to remind users within the ad campaign list. For example, ... Figure 3 As shown, it displays a user's ad campaign list interface in an example. The "Adjustable Volume" message at position 31 is a reminder provided by the "Marketing Assistant". Users can use this reminder to discover that their campaign has room for optimization, which can then guide them into the optimization process.
[0075] Additionally, an inspection function can be provided. Specifically, users can choose whether to enable the inspection mode. If enabled, real-time inspections of specific advertising campaigns can be conducted. Inspection reports can then guide users on how to express their specific support needs to enter optimization or in-depth diagnostic modes. The inspection process does not necessarily rely on AI models; it can be achieved through data statistics and other methods to calculate the performance data of a specific advertising campaign across multiple inspection items. For example, specific inspection items could include acquisition volume, cost of goods sold, spending rate, and performance analysis.
[0076] Furthermore, users can also input dialogue content for intelligent diagnosis of the target advertising campaign through the aforementioned dialogue entry. At this time, the AI model can understand the received dialogue content and determine the user's intent, analyze the problems existing in the target advertising campaign during the campaign and the corresponding reasons, and generate diagnostic results. Users can express their optimization needs based on the diagnostic results.
[0077] It's important to note that the diagnosis described here differs from the inspection mentioned earlier. Inspections typically only involve data statistics and can yield preliminary results without the aid of AI models. However, diagnosis requires AI models. Diagnostic services can not only uncover potential problems in specific advertising campaigns but also enable root cause analysis, representing a more in-depth form of diagnosis.
[0078] In an optional mode, after a user initiates a diagnosis, the specific diagnostic results can not only display the specific problems existing in the current plan and provide corresponding cause analysis results, but also generate content to guide the user in optimizing the target advertising campaign. Furthermore, the AI model generates suggested information on the optimization direction of the target advertising campaign. For example, it may suggest budget optimization, creative optimization, etc. Additionally, in this embodiment, algorithm optimization is also supported, that is, optimizing the processing elements in the intelligent advertising algorithm in the background, as mentioned above. This guidance information can guide users to use the optimization function.
[0079] S202: The AI module is used to understand the dialogue content to determine the user's intent. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, the reasonableness of the user's intent is judged.
[0080] Since diagnosis and optimization can share the same dialogue entry point, in practice, user intent recognition can be performed after receiving the user's dialogue content. Specifically, separate AI diagnosis and AI optimization modules can be provided, along with an AI traffic routing module, and so on. The AI traffic routing module and AI diagnosis module can be independent of each other, or they can be performed by the same AI module. These AI traffic routing, AI diagnosis, and AI optimization modules can also be referred to as a traffic routing agent, a diagnosis agent, and an optimization agent, respectively.
[0081] Specifically, the AI-driven traffic routing module can first determine whether a user needs diagnostics or optimization of their ad campaign. If the user needs diagnostics, they can be routed to the AI diagnostic module; otherwise, if they need optimization, they can be routed to the AI optimization module. When the user's intent is optimization, this involves adjusting specific elements processed by the intelligent ad delivery algorithm, including blocking / emphasizing specific keywords / audiences, adjusting bid coefficients, etc. Such optimization might result in the user's expressed optimization request affecting the platform's overall goals. Therefore, in implementation, the reasonableness of the user's intent can be assessed. For example, suppose a user wants to increase the bid coefficient by a certain value, but this adjustment will affect the achievement of the overall goals. Therefore, the user can be prompted to modify the specific value and then re-express their intent, and so on.
[0082] Specifically, when expressing specific optimization needs, users can control the intelligent targeting algorithm to either block or prioritize specific keywords, audiences, or targeting areas when intelligently selecting keywords, audiences, or targeting zones. Alternatively, users can control the intelligent targeting algorithm to adjust the bid coefficients for specific keywords / audiences or time periods during intelligent bidding. Furthermore, users can express multiple processing elements separately, such as, "I want to increase the spend rate while adjusting traffic in a certain targeting zone and setting the targeting to ***," etc.
[0083] S203: If the rationality judgment is passed, the AI module determines the optimization scheme corresponding to the user's intent, and optimizes the target information delivery plan according to the optimization scheme.
[0084] After identifying the user's intent as optimization and passing the reasonableness judgment, the AI module can determine the optimization plan corresponding to the user's intent and optimize the target advertising campaign according to the optimization plan. Specifically, the user's expression may be vague, such as simply "I want to increase the spending rate" or "adjust the traffic in a certain advertising area." However, when executing the optimization plan, it is necessary to determine the specific amount to adjust the spending rate, or how to adjust the traffic in a certain advertising area, etc. Therefore, after receiving the user's dialogue, the AI module can also generate an optimization plan. Specifically, when generating the optimization plan, the user's expressed optimization needs can be broken down first, generating multiple optimization tasks with the AI module as the main body. For example, for the user's expression in the aforementioned example, "I want to increase the spending rate, adjust the traffic in a certain advertising area, and set the targeting to ***," the AI module's breakdown of the optimization steps could be as follows: Figure 4 As shown, this can be broken down into the following optimization tasks:
[0085] Targeted planning: I need to change the target audience of plan ID***** to ******, and the system will automatically help them explore customers;
[0086] Traffic control: I need to set the traffic targeting preference for the plan ID***** to ***, and focus on targeting the *** and *** domains;
[0087] Time-based spending control: I want to help clients get more traffic between 0:00 and 6:00, and I expect spending to be higher during this period.
[0088] After breaking down the optimization tasks as described above, the AI module can execute each of these tasks individually. The AI tuning module can be further divided into multiple sub-modules, each corresponding to the tuning of different processing object elements. For example, it could include keyword agents, audience (targeting) agents, bidding agents, and traffic agents, etc. After the tuning agents have completed the decomposition of optimization tasks, they can also assign specific optimization tasks to the sub-agents corresponding to the specific processing object elements. For example, the "targeting planning" task in the above example can be assigned to the audience (targeting) agent, the "traffic control" task can be assigned to the traffic agent, the "time-based spending control" task can be assigned to the bidding agent, and so on.
[0089] After the specific AI optimization module or sub-module completes its optimization task, that is, the target advertising campaign is optimized according to the specific optimization plan. In other words, in the solution provided in this application embodiment, the merchant user initiates an optimization request for a specific advertising campaign by inputting dialogue content. Then, the AI module can complete the optimization process. For example, if the user needs to focus more on advertising a certain keyword, the AI module's processing result is to advertise that keyword with a higher weight, and so on, thereby achieving an "end-to-end" solution to the customer's marketing needs.
[0090] In other words, this application embodiment can use AI from the platform's perspective to agent-ize the core capabilities of advertising placement, such as keywords, audience (targeting), and bidding, based on the client's planned placement needs. It breaks down the client's needs in natural language expression "end-to-end" and modifies the underlying placement engine and engineering to achieve intelligent placement based on the client's self-expressed goals, thus controlling the achievement of client objectives. AI also provides a more comprehensive understanding of client needs, more efficient and consistent decision-making across the entire process, and more intelligent data analysis and reflective adjustments, forming a "one-customer-one-policy" delivery capability and comprehensively improving client satisfaction.
[0091] It should be noted that each Agent in the embodiments of this application can be pre-trained based on industry data, expert experience data, etc., so that it has the ability to recognize intent, diagnose, and optimize.
[0092] It's also worth noting that in practice, after a user has optimized a particular advertising campaign, they can add new requests. For example, a previous request might have been "I want to acquire traffic and have it targeted broad category keywords," and optimizations might have already been completed through collaboration among the various agents. However, the user might later add the following request: "I also want to target female audiences and delete the traffic acquisition target." In this case, the AI module can first search for historical requests and combine them with the new request to arrive at the user's latest request: "The client wants to target broad category keywords and specifically targets female audiences." Then, based on this latest request, tasks can be broken down and executed to complete the specific optimizations.
[0093] In other words, for specific advertising campaigns, merchants can adjust their campaign needs at any time based on the actual situation. When adjustments are needed, they only need to express their specific campaign needs again through the dialogue portal, including expressing which one or more processing elements in the intelligent campaign algorithm need to be adjusted, and even specifying how to adjust them, etc.
[0094] In summary, through the embodiments of this application, for information delivery plans that are partially or entirely executed through intelligent delivery algorithms, users can submit dialogue content for communication with the AI module during the plan execution process. Optimization can be achieved by intervening in the processing elements of the intelligent delivery algorithm. After receiving the user's dialogue content, if the AI module identifies the aforementioned intent, it can determine an optimization scheme corresponding to the user's intent after a reasonableness assessment, and optimize the target information delivery plan according to the optimization scheme to achieve an "end-to-end" solution to customer marketing needs. Through this solution, intelligent delivery algorithms, which were previously completely "black box" in nature, become operable, and users can express specific optimization requests in a "white box" manner. This addresses problems in the delivery process from a more fundamental perspective, allowing information delivery plans to be tailored to the specific needs of each merchant, rather than being limited to a few selectable options. Because optimization plans can be generated based on the "white-box" dialogue content of merchants and users, more merchant preference data can be expressed. Therefore, the intelligent delivery algorithm involved in the information delivery plan becomes a kind of "tunable intelligence".
[0095] Example 2
[0096] In the aforementioned Embodiment 1, the specific solution was mainly introduced from the perspective of optimization. However, the diagnostic function is also a very important aspect, and as described in Embodiment 1, the diagnostic function and the optimization function can share the same dialog entry point. Therefore, Embodiment 2 mainly addresses the situation where the diagnostic function and the optimization function share the same dialog entry point, and provides an advertising campaign processing system. See [link to Embodiment 1]. Figure 5 The method may specifically include:
[0097] The dialogue module 501 is used to receive dialogue content submitted by the user regarding the target information delivery plan during the execution of the target information delivery plan; the target information delivery plan is executed intelligently, partly or entirely, through an intelligent delivery algorithm.
[0098] AI traffic routing module 502 is used to understand the dialogue content to determine the user intent. If the user intent is to diagnose the target information delivery plan, the traffic is routed to the AI diagnosis module. If the user intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, the user intent is judged for reasonableness. If the reasonableness judgment is passed, the traffic is routed to the AI optimization module.
[0099] The AI diagnostic module 503 is used to analyze the problems and corresponding causes that exist in the target information delivery plan during the delivery process, and generate diagnostic results.
[0100] The AI optimization module 504 is used to determine an optimization scheme corresponding to the user intent, and optimize the target information delivery plan according to the optimization scheme.
[0101] The AI diagnostic module is also used to generate content to guide users in optimizing the target information delivery plan.
[0102] In practice, the AI traffic splitting module and the AI diagnostic module are the same module. Of course, the AI traffic splitting module and the AI diagnostic module can also be different modules, depending on the actual needs.
[0103] The AI optimization module may include multiple AI optimization sub-modules, each corresponding to a different processing object element;
[0104] The AI tuning module is also used to break down the user intent into multiple optimization tasks, each optimization task corresponding to a different processing object element, and to assign the optimization task to the corresponding AI tuning sub-module according to the processing object element information, so as to generate optimization schemes and perform optimization processing.
[0105] For the parts of this embodiment that are not described in detail, please refer to the description in the foregoing embodiment one and other parts of this specification, which will not be repeated here.
[0106] 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.).
[0107] Corresponding to Embodiment 1, this application also provides an information delivery plan processing device, which may include:
[0108] The dialogue content receiving unit is used to receive dialogue content submitted by the user in response to the target information delivery plan during the execution of the target information delivery plan; the target information delivery plan is executed partially or entirely through intelligent delivery algorithms.
[0109] The intent understanding unit is used to understand the dialogue content through the artificial intelligence (AI) module to determine the user's intent. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, then the reasonableness of the user's intent is judged.
[0110] The optimization unit is used to determine the optimization scheme corresponding to the user's intent through the AI module when the rationality judgment is passed, and to optimize the target information delivery plan according to the optimization scheme.
[0111] Specifically, the dialogue content receiving unit can be used for:
[0112] During the execution of the target information delivery plan, a dialogue entry point is provided for communicating with the AI module, so as to receive user-inputted dialogue content optimized for the target information delivery plan through the dialogue entry point.
[0113] In a specific implementation, the device may further include:
[0114] The inspection report providing unit is used to provide an inspection report on the target information delivery plan before the user inputs dialogue content. The inspection report includes data on the performance of the target information delivery plan on multiple inspection items, so that the user can express optimization needs based on the inspection results.
[0115] In addition, the dialogue content receiving unit can also be used to receive dialogue content input by the user for intelligent diagnosis of the target information delivery plan through the dialogue entry point;
[0116] The intent understanding unit can also be used to understand the received dialogue content and determine the user's intent through an AI model, analyze the problems and corresponding reasons in the delivery process of the target information delivery plan, and generate diagnostic results so that the user can express their optimization needs based on the diagnostic results.
[0117] Furthermore, the device may also include:
[0118] The content provision unit is used to generate content through the AI model to guide users in optimizing the target information delivery plan.
[0119] The suggestion information providing unit is used to generate suggestion information on the optimization direction of the target information delivery plan through the AI model.
[0120] Specifically, the intervention on the processing object elements in the intelligent delivery algorithm includes:
[0121] The intelligent targeting algorithm can be controlled to either block or emphasize specific keywords, audiences, or target areas when intelligently selecting keywords, audiences, or target areas.
[0122] Alternatively, the intelligent bidding algorithm can be controlled to adjust the bid coefficient for specified keywords / audiences or time periods during intelligent bidding.
[0123] 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.
[0124] And an electronic device, comprising:
[0125] One or more processors; and
[0126] 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.
[0127] 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.
[0128] in, Figure 6An exemplary architecture of an electronic device is shown, which may include a processor 610, a video display adapter 611, a disk drive 612, an input / output interface 613, a network interface 614, and a memory 620. The processor 610, video display adapter 611, disk drive 612, input / output interface 613, network interface 614, and memory 620 can communicate with each other via a communication bus 630.
[0129] The processor 610 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.
[0130] The memory 620 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 620 can store the operating system 621 for controlling the operation of the electronic device 600, and the basic input / output system (BIOS) for controlling the low-level operations of the electronic device 600. Additionally, it can store a web browser 623, a data storage management system 624, and an information delivery plan processing system 625, etc. The aforementioned information delivery plan processing system 625 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 620 and executed by the processor 610.
[0131] Input / output interface 613 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.
[0132] Network interface 614 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0133] Bus 630 includes a pathway for transmitting information between various components of the device, such as processor 610, video display adapter 611, disk drive 612, input / output interface 613, network interface 614, and memory 620.
[0134] It should be noted that although the above-described device only shows the processor 610, video display adapter 611, disk drive 612, input / output interface 613, network interface 614, memory 620, bus 630, 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.
[0135] 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.
[0136] 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.
[0137] The foregoing has provided a detailed description of the information delivery plan processing method, system, and electronic equipment provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for processing information delivery plans, characterized in that, include: During the execution of the target information delivery plan, the system receives dialogue content submitted by users regarding the target information delivery plan. The target information delivery plan is executed partially or entirely through intelligent delivery algorithms. The AI module is used to understand the dialogue content to determine the user's intent. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, then the reasonableness of the user's intent is judged. If the rationality assessment is passed, the AI module determines the optimization scheme corresponding to the user's intent, and optimizes the target information delivery plan according to the optimization scheme.
2. The method according to claim 1, characterized in that, The process of receiving dialogue content submitted by the user regarding the target information delivery plan includes: During the execution of the target information delivery plan, a dialogue entry point is provided for communicating with the AI module, so as to receive user-inputted dialogue content optimized for the target information delivery plan through the dialogue entry point.
3. The method according to claim 2, characterized in that, Also includes: Before the user inputs dialogue content, an inspection report on the target information delivery plan is provided. The inspection report includes data on the performance of the target information delivery plan on multiple inspection items, so that the user can express optimization needs based on the inspection results.
4. The method according to claim 2, characterized in that, Also includes: The dialogue entry point receives user input regarding the intelligent diagnosis of the target information delivery plan. After understanding the received dialogue content and determining the user's intent through an AI model, the system analyzes the problems and corresponding reasons in the delivery process of the target information delivery plan, generates diagnostic results, and allows users to express their optimization needs based on the diagnostic results.
5. The method according to claim 4, characterized in that, Also includes: The AI model generates content to guide users in optimizing the target information delivery plan.
6. The method according to claim 4, characterized in that, Also includes: The AI model generates suggested directions for optimizing the target information delivery plan.
7. The method according to claim 1, characterized in that, The intervention on the processing object elements in the intelligent delivery algorithm includes: The intelligent targeting algorithm can be controlled to either block or emphasize specific keywords, audiences, or target areas when intelligently selecting keywords, audiences, or target areas.
8. The method according to claim 1, characterized in that, The intervention on the processing object elements in the intelligent delivery algorithm includes: The intelligent bidding algorithm adjusts the bid coefficients for specified keywords / audiences or time periods during intelligent bidding.
9. An information delivery plan processing system, characterized in that, include: The dialogue module is used to receive dialogue content submitted by users regarding the target information delivery plan during the execution of the target information delivery plan; The target information delivery plan is executed partially or entirely through intelligent delivery algorithms, and the dialogue content is used to express the needs generated during the execution of the target information delivery plan. The AI traffic routing module is used to understand the dialogue content to determine the user's intent. If the user's intent is to diagnose the target information delivery plan, the traffic is routed to the AI diagnosis module. If the user's intent is to optimize by intervening in the processing object elements in the intelligent delivery algorithm, the user's intent is judged for reasonableness. If the reasonableness judgment is passed, the traffic is routed to the AI optimization module. The AI diagnostic module is used to analyze the problems and corresponding causes that exist in the target information delivery plan during the delivery process, and generate diagnostic results. The AI optimization module is used to determine the optimization scheme corresponding to the user intent, and optimize the target information delivery plan according to the optimization scheme.
10. The system according to claim 9, characterized in that, The AI diagnostic module is also used to generate content to guide users in optimizing the target information delivery plan.
11. The system according to claim 9, characterized in that, The AI triage module and the AI diagnostic module are the same module.
12. The system according to claim 9, characterized in that, The AI tuning module includes multiple AI tuning sub-modules, each corresponding to a different processing object element; The AI tuning module is also used to break down the user intent into multiple optimization tasks, each optimization task corresponding to a different processing object element, and to assign the optimization task to the corresponding AI tuning sub-module according to the processing object element information, so as to generate optimization schemes and perform optimization processing.
13. 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.
14. 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.
15. 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.