An advertisement delivery method and device, electronic equipment and storage medium

Abstract

The application discloses an advertisement putting method and device, electronic equipment and computer readable storage medium. The embodiment of the application obtains target available resource quantity of multiple advertisement positions and expected income index values of the advertisement positions. Based on the target available resource quantity and the expected income index values of the advertisement positions, at least two advertisement positions with complementary resource and income attributes are combined and associated to obtain at least one advertisement position combination. The advertisement position combination includes at least two advertisement positions with different resource and income attributes. The application obtains advertisement putting indexes of a target advertisement to be put. Based on the advertisement putting indexes, a target advertisement position combination is selected from the advertisement position combination, and the target advertisement is put into the target advertisement position combination. The application can reduce the data processing pressure and system resource consumption of a server when responding to high-concurrency advertisement requests, and can improve the overall resource utilization of the system.

Description

Technical Field

[0001] This disclosure relates to the field of advertising delivery technology, specifically to an advertising delivery method, apparatus, electronic device, and storage medium. Background Technology

[0002] In related technologies, servers typically maintain each ad slot as an independent data management unit. Because the inventory data, bidding parameters, and delivery strategies for each ad slot are fragmented, servers struggle to uniformly index and schedule multi-source heterogeneous data when responding to high-concurrency ad requests. This results in a large amount of repetitive calculations and redundant signaling interactions, significantly increasing data processing pressure and system resource overhead. Furthermore, the delivery progress data for each ad slot is usually stored independently, preventing servers from synchronizing and aligning the delivery status across multiple dimensions globally in real time, further limiting the overall resource utilization efficiency of the system. Summary of the Invention

[0003] This application provides an advertising delivery method, apparatus, electronic device, and storage medium, which can reduce the data processing pressure and system resource overhead of the server when responding to high-concurrency advertising requests, and improve the overall resource utilization of the system.

[0004] In a first aspect, embodiments of this application provide an advertising delivery method, the method comprising: Obtain the target available resources for multiple ad placements, and the expected revenue metrics for each ad placement; Based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination. The ad slot combination includes at least two ad slots with different resource and revenue attributes. Obtain the advertising metrics for the target ads to be delivered; Based on advertising performance metrics, select target ad placement combinations from ad placement combinations and deliver target ads to these target ad placement combinations.

[0005] Secondly, embodiments of this application provide an advertising delivery device, the device comprising: The information acquisition module is used to acquire the target available resources for multiple ad slots, as well as the expected revenue metrics for each ad slot. The ad placement combination module is used to combine and associate at least two ad placements with complementary resource and revenue attributes based on the target available resource amount and expected revenue index value of each ad placement, so as to obtain at least one ad placement combination. The ad placement combination includes at least two ad placements with different resource and revenue attributes. The metrics acquisition module is used to acquire the advertising metrics for the target ads to be delivered. The ad delivery module is used to select target ad placement combinations from ad placement combinations based on ad delivery metrics, and to deliver target ads to the target ad placement combinations.

[0006] Thirdly, embodiments of this application also provide an electronic device, including a memory storing multiple instructions; a processor loads instructions from the memory to execute the steps of any of the advertising delivery methods provided in embodiments of this application.

[0007] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a plurality of instructions adapted for loading by a processor to execute the steps of any of the advertising delivery methods provided in embodiments of this application.

[0008] Fifthly, embodiments of this application also provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps in any of the advertising delivery methods provided in embodiments of this application.

[0009] The solution adopted in this application embodiment can obtain the target available resource quantity of multiple ad slots and the expected revenue index value of each ad slot; based on the target available resource quantity and expected revenue index value of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, the ad slot combination including at least two ad slots with different resource and revenue attributes; obtain the advertising delivery index of the target advertisement to be delivered; based on the advertising delivery index, select the target ad slot combination from the ad slot combination and deliver the target advertisement to the target ad slot combination. In this way, by using the ad slot combination method, the server can perform unified indexing and scheduling of multi-source heterogeneous data when responding to high-concurrency advertising requests, avoiding a large amount of repetitive calculation and redundant signaling interaction, thereby greatly reducing data processing pressure and system resource consumption. Furthermore, it can enable the server to synchronize and align the delivery status of multiple dimensions in real time at the global level, so as to improve the overall resource utilization of the system. Attached Figure Description To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic flowchart of one embodiment of the advertising delivery method provided in this application. Figure 2 This is a schematic diagram of the advertising delivery system provided in the embodiments of this application; Figure 3 This is a schematic flowchart of another embodiment of the advertising delivery method provided in this application. Figure 4 This is a schematic diagram of the advertising delivery device provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0011] 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 a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. At the same time, in the description of the embodiments of this application, the terms "first," "second," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0012] This application provides an advertising delivery method, apparatus, electronic device, and computer-readable storage medium.

[0013] Specifically, this embodiment will be described from the perspective of an advertising delivery device, which can be integrated into an electronic device. That is, the advertising delivery method of this embodiment can be executed by an electronic device. Optionally, the electronic device may include a terminal device. The terminal device may be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, game console, or personal computer (PC), etc.

[0014] The advertising delivery method provided in this application can be applied to advertising delivery systems. These systems can include player terminal devices and servers. The terminal can be a device that includes both receiving and transmitting hardware, i.e., a device with receiving and transmitting hardware capable of performing bidirectional communication over a bidirectional communication link. The player terminal device and the server can communicate bidirectionally via a network.

[0015] Optionally, the server can be a standalone server, or a server network or server cluster, including but not limited to computers, network hosts, single network servers, multiple network server sets, or cloud servers composed of multiple servers. Cloud servers consist of a large number of computers or network servers based on cloud computing.

[0016] The following detailed description is provided in conjunction with the accompanying drawings. In this embodiment, the execution subject is a terminal device as an example. It should be noted that the order of description in the following embodiments is not intended to limit the preferred order of the embodiments. Although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be performed in a different order than that shown in the accompanying drawings.

[0017] The advertising delivery method in this embodiment can reduce the data processing pressure and system resource consumption of the server when responding to high-concurrency advertising requests, and can improve the overall resource utilization of the system.

[0018] Please refer to Figure 1 Taking an advertising delivery system as an example, this embodiment provides an advertising delivery method. The specific process of this advertising delivery method can be summarized in steps 101 to 105, wherein: Step 101: Obtain the target available resources for multiple ad slots, as well as the expected revenue metrics for each ad slot.

[0019] The aforementioned advertising space refers to a specific area or time slot reserved on various media platforms (such as websites, apps, social media, outdoor screens, etc.) specifically for displaying commercial advertising content. This advertising space can present specific advertisements in the form of images, text, videos, audio, etc.

[0020] In this embodiment, in order to better evaluate each ad slot and to facilitate subsequent processing of the ad slots based on the evaluation results, the ad delivery system can introduce target available resources and expected revenue indicators to characterize the evaluation results of the ad slots.

[0021] The target available resources refer to the amount of resources that can be used by the ad placement within a specific period. The target available resources can indicate the inventory of the ad placement within a specific period. The target available resources may include, but are not limited to, available impressions, available traffic requests, available ad display duration, and targeted audience packages. The specific settings can be configured according to needs and are not limited here.

[0022] The expected revenue indicator is the revenue that the ad placement is expected to obtain within a specific period after the ad campaign. This target available resource quantity can be used to measure the efficiency of the ad placement combination after the ad campaign within a specific period. This target available resource quantity may include, but is not limited to, click-through rate (CTR), effective cost per mille (eCPM), conversion rate (such as CVR), etc. The specific value can be set according to the needs and is not limited here.

[0023] Click-through rate (CTR) refers to the ratio of the number of times an ad is clicked to the number of times it is displayed on an ad slot.

[0024] Specifically, advertising delivery systems can use inventory forecasting to estimate the inventory demand for ad slots in a specific future time period. For example, they can forecast the exposure of ad slots in a specific future time period and use the forecast results as the target available resource quantity.

[0025] In some embodiments, the advertising delivery system can obtain historical ad space data for a historical period, such as data from the past 30 days. The specific historical period for data collection can be set according to requirements and is not limited here. Then, feature extraction is performed based on the historical ad space data to obtain historical ad space features. Finally, the target available resources and expected revenue indicators of the ad space are predicted based on the historical ad space features.

[0026] Historical ad placement data may include, but is not limited to, historical exposure data, click data, and user behavior data.

[0027] Accordingly, the characteristics of historical ad placements may include, but are not limited to, time characteristics (hours, days of the week, holidays), user characteristics (region, device, audience tags), and contextual characteristics (page type, content category).

[0028] Specifically, predicting the target available resource quantity based on historical ad slot characteristics may include: inputting historical ad slot characteristics into a time series prediction model (e.g., an XGBoost model) to predict the target available resource quantity through the time prediction model.

[0029] Optionally, in order to more accurately predict the target available resources and perform subsequent processing based on the prediction results, the target available resources can be predicted in a time-sharing manner, that is, the available resources of each sub-time period in the future time period can be predicted. For example, the target available resources (such as available exposure) of the ad slot in each time period in the next 24 hours can be predicted, such as predicting in 15-minute time periods (96 time periods in total in 24 hours). The prediction result is expressed as the estimated exposure vector of each time period.

[0030] For example, the formula for predicting the amount of available resources for a target is as follows: Inventory_{t} = f(X_{t-1}, X_{t-2}, ..., X_{tn}) + ε.

[0031] Where X_{t-1} represents the historical ad slot features for the time period t-1, X_{t-2} represents the historical ad slot features for the time period t-2, X_{tn} represents the historical ad slot features for the time period tn, ε represents the error correction term, f represents the prediction model, and Inventory_{t} represents the target available resource quantity.

[0032] Specifically, predicting expected revenue metrics based on historical ad placement characteristics may include: inputting historical ad placement characteristics into a machine learning model to predict expected revenue metrics through the machine learning model.

[0033] It should be noted that, since the machine learning model needs to consider the ad placement combination effect when predicting expected revenue indicators, the historical ad placement features input into the machine learning model can include historical efficiency metrics of sub-ad placements, historical inventory (such as available exposures), and combination features. These combination features can be the cross-embedding of different ad placements with tag identifiers (which can be targeting condition identifiers, such as targeting a group), physical distance features, etc. The specific settings can be configured according to the requirements and are not limited here.

[0034] The labels can be targeted at specific conditions, such as the gender or age group of the target audience. The specific labels can be set according to the needs and are not limited here.

[0035] For example, machine learning models can use DNN models.

[0036] For example, such as Figure 2 As shown, the advertising delivery system can be equipped with a traffic forecasting module (denoted as InventoryForecasting). The advertising delivery system can use this traffic forecasting module to determine the target available resources and expected revenue metrics for the ad slots.

[0037] Step 102: Based on the target available resources and expected revenue indicators of each ad slot, combine and associate at least two ad slots with complementary resource and revenue attributes to obtain at least one ad slot combination. The ad slot combination includes at least two ad slots with different resource and revenue attributes.

[0038] In this embodiment, the advertising delivery system estimates the traffic of the ad slots in real time. Based on two dimensions of information—the target available resources and expected revenue indicators for the ad slots—the system can combine ad slots, that is, combine and associate at least two ad slots with complementary resource and revenue attributes to obtain at least one ad slot combination.

[0039] Understandably, complementary resources indicate that ad placements are complementary at the resource level. That is, ad placements that are complementary at the resource level are identified by the target available resource amount. For example, if one ad placement has a large target available resource amount, while another ad placement has a small target available resource amount, the two ad placements are combined to make the two ad placements with different target available resource amounts complementary at the resource level.

[0040] It's understandable that ad placements with higher revenue attributes may have fewer remaining target available resources, while ad placements with lower revenue attributes may have more remaining target available resources. This can sometimes lead to situations where the target available resources for ad placements with higher revenue attributes cannot meet user demand, while the revenue attributes of ad placements with more target available resources cannot meet user demand. Consequently, some ad placements may be left unused during the campaign. Therefore, this situation can be avoided by combining ad placements with different revenue attributes. For example, one ad placement may have a higher expected revenue metric than another with a lower expected revenue metric. Combining these two ad placements can create a complementary relationship in terms of revenue attributes.

[0041] In some embodiments, the advertising delivery system can dynamically classify ad slots based on two dimensions: the target available resources and the expected revenue index value for each ad slot. This will divide the ad slots into corresponding ad categories. Then, based on the ad categories of each ad slot, the system can combine the ad slots to obtain at least one ad slot combination. This ad slot combination may include ad slots from at least two ad categories. This clear category helps to accurately target ads and improve the dissemination effect. Furthermore, by combining ad slots from different ad categories, the system can combine ad slots with different characteristics to obtain multiple ad slot combinations, thereby achieving a balance of characteristics between different ad slot combinations and realizing the dynamic integration of multiple ad slots to make advertising delivery decisions.

[0042] Specifically, the classification of ad categories for ad slots may include, but is not limited to: comparing the target available resources and expected revenue indicators of an ad slot with a preset threshold, and using the category corresponding to the preset threshold range that the target available resources and expected revenue indicators of an ad slot meet as the ad category of that ad slot; or, inputting the target available resources and expected revenue indicators of an ad slot into a preset neural network model, so that the neural network model can perform category identification based on the target available resources and expected revenue indicators of the ad slot to obtain the ad category of that ad slot.

[0043] It should be noted that the above-mentioned advertising categories are used to indicate the characteristics of the advertising space in a specific dimension. In this embodiment, since the above-mentioned advertising categories are obtained based on the target available resources and expected revenue indicators of the advertising space, the above-mentioned advertising categories are used to indicate the characteristics of the advertising space in the resource usage dimension and the revenue dimension.

[0044] In some embodiments, the above-mentioned advertising categories may include at least one of a first category, a second category, and a third category, and different categories may be divided for advertising spaces based on the target available resources and expected revenue indicators of the advertising space.

[0045] Among them, the target available resources of the first category (denoted as L-type) of the ad space are greater than the first preset resource threshold, and the expected revenue index value is less than the first preset revenue threshold. That is, the L-type ad space belongs to the low-efficiency, high-inventory type of ad space.

[0046] Among them, the target available resources of the second category (denoted as H type) of the ad space are less than the second preset resource threshold, and the expected revenue index value is greater than the second preset revenue threshold. That is, the H type ad space belongs to the high-efficiency low-inventory type of ad space.

[0047] Among them, the target available resources of the third category (denoted as type B) of the ad space are less than or equal to the first preset resource threshold and greater than or equal to the second preset resource threshold, and the expected revenue index value is greater than or equal to the first preset revenue threshold and less than or equal to the second preset revenue threshold. That is, type B ad space belongs to the balanced ad space.

[0048] Wherein, the first preset resource threshold is greater than or equal to the second preset resource threshold, and the first preset revenue threshold is less than or equal to the second preset revenue threshold.

[0049] For example, when the expected revenue metric is click-through rate, the first preset revenue threshold can be set to 0.02 and the second preset revenue threshold can be set to 0.05; when the target available resources are measured in page views (PV), the first preset resource threshold can be set to 10,000 and the second preset resource threshold can be set to 50,000.

[0050] Specifically, the advertising delivery system can first dynamically pair different types of ad slots in a ratio (e.g., N:M), and then automatically perform ad slot fusion (called Slot Fusion) based on the pairing results to create at least one ad slot combination, thus avoiding the manpower costs required for manual configuration.

[0051] Among them, ad placement fusion refers to the strategy of logically merging multiple physical ad placements into a unified delivery unit. In this embodiment, ad placements can be paired in a many-to-many ratio according to an N:M ratio to create at least one ad placement combination.

[0052] Understandably, by dynamically combining different types of ad placements (such as high-efficiency low-inventory and low-efficiency high-inventory), the advantages of different ad placements can be effectively integrated. Furthermore, there is no need to set up separate placement strategies for each ad placement, which reduces the operating costs of ad placement. There is also no need to maintain the inventory and pricing of each ad placement separately, which reduces the complexity of the system.

[0053] In some embodiments, the ad placement mix may include at least ad placements of the first and second categories to dynamically combine high-efficiency, low-inventory ad placements with low-efficiency, high-inventory ad placements. This addresses the current problem of resource fragmentation, where different ad placements, due to factors such as location, size, and target audience, exhibit significant differences in click-through rate (CTR) and exposure inventory. Consequently, high-efficiency ad placements face intense competition and high bidding costs due to limited inventory, while low-efficiency ad placements are neglected due to poor performance, resulting in ample inventory that is difficult to sell. This leads to wasted media resources and low ad placement efficiency, preventing the ad placements from complementing each other and resulting in low overall resource utilization.

[0054] It should be noted that since the third category of ad slots is a balanced ad slot, during the ad placement process, the third category of ad slots will generally be allocated first, while the remaining slots will be specific ad slots, namely the first and second categories of ad slots. Therefore, in this embodiment, by combining the first and second categories of ad slots, the resource utilization rate can be improved and the current resource fragmentation problem can be solved.

[0055] In some embodiments, based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, including: determining the ad category of each ad slot based on the target available resources and expected revenue indicators of each ad slot; and, under the condition of satisfying preset ad slot combination constraints, combining and associating at least two ad slots with complementary resource and revenue attributes based on the ad category of each ad slot to obtain at least one ad slot combination. The preset ad placement combination constraints may include at least one of the following: The expected combined revenue index value corresponding to the ad placement combination is greater than or equal to the preset combined revenue index threshold. The available resource amount of the target combination corresponding to the ad placement combination is greater than or equal to the preset threshold of available resource amount for the combination. The audience overlap of each ad placement in the ad placement combination is less than the preset overlap threshold.

[0056] In this embodiment, the advertising delivery system uses a combination of factors, such as efficiency constraints based on combined revenue metrics, inventory constraints based on available resources, and audience overlap constraints based on audience overlap, to automatically pair ad slots and create at least one ad slot combination. For example, the ad categories of ad slots can be abstracted as "high-efficiency, low-inventory" and "low-efficiency, high-inventory". Then, intelligent fusion and pairing between ad slots is achieved through three constraints, thereby covering one-to-one and many-to-many scenarios.

[0057] Understandably, compared to managing each ad slot independently, which fails to effectively integrate ad slot resources with different advantages, this embodiment constructs an ad slot classification framework based on the two-dimensional ad slot information of the target available resource quantity and expected revenue index value of the ad slot. It also uses preset ad slot combination constraints to jointly constrain the construction process of ad slot combination, so as to ensure the integration quality of ad slot combination and achieve resource integration with complementary advantages.

[0058] Specifically, the expected combined revenue index value corresponding to the ad placement combination can be obtained by weighting and averaging the expected combined revenue index values ​​of each ad placement in the ad placement combination, or by taking the maximum value, median value, etc., which can be set according to the needs and are not limited here.

[0059] Accordingly, the preset combined revenue indicator threshold can be determined based on the ad placement combination. For example, the expected revenue indicator value of an ad placement within the ad placement combination (such as the expected revenue indicator value of an H-type ad placement to ensure that the placement efficiency does not drop significantly) can be selected as the first intermediate value. Alternatively, the expected revenue indicator values ​​of ad placements of a specific category (such as H-type) within the ad placement combination can be weighted and averaged or calculated to obtain the first intermediate value. Then, the product between the first intermediate value and the first preset coefficient (denoted as α) can be calculated as the preset combined revenue indicator threshold.

[0060] Specifically, the available resource quantity of the target combination corresponding to the ad placement combination can be obtained by weighted averaging of the target available resource quantities of each ad placement in the ad placement combination, or by taking the maximum value, median value, etc., which can be set according to the needs and are not limited here.

[0061] Accordingly, the preset available resource threshold can be determined based on the ad placement combination. For example, the target available resource of an ad placement (such as the target available resource of an L-shaped ad placement to ensure inventory utilization) can be selected as the second intermediate value. Alternatively, the target available resource of a specific category (such as an L-shaped ad placement) within the ad placement combination can be weighted and averaged or calculated to obtain the second intermediate value. Then, the product between the second intermediate value and the second preset coefficient (denoted as β) can be calculated as the preset available resource threshold.

[0062] Specifically, audience overlap can be calculated by determining the similarity between the audiences of each ad placement (such as the Jaccard similarity algorithm).

[0063] The values ​​of the preset overlap threshold (denoted as γ), α, and β can be set according to requirements and are not limited here. For example, α=0.8, β=0.9, γ=0.3.

[0064] In some embodiments, after obtaining at least one ad slot combination (such as denoted as Fused Slot), a uniform identifier can be assigned, that is, a mapping relationship between different ad slot combinations and different identifiers can be established, that is, a mapping relationship between multiple ad slots within an ad slot combination and their corresponding identifiers can be established.

[0065] Optionally, since different ad placements have different targeting conditions (such as the target user groups of each ad placement), an ad placement combination can inherit the targeting conditions of multiple ad placements within the ad placement combination, that is, the union of the targeting conditions of multiple ad placements within the ad placement combination is used as the targeting conditions corresponding to the ad placement combination.

[0066] Optionally, since different ad slots have different bid floor prices, the bid floor price of an ad slot combination can be the weighted average or average value of multiple ad slots in the ad slot combination, etc. The specific setting can be determined according to the needs and is not limited here.

[0067] The bid floor price can be obtained through real-time bidding (RTB), which refers to the mechanism by which ad placements complete bidding and placement decisions within milliseconds.

[0068] For example, such as Figure 2 As shown, the advertising delivery system can be equipped with a fusion decision module (denoted as SlotFusion Decision). The advertising delivery system can determine the advertising category of the ad slot based on the fusion decision module, and combine the ad slots based on the determined advertising category to obtain at least one virtual ad combination.

[0069] Step 104: Obtain the advertising metrics for the target ads to be placed.

[0070] The advertising metrics may include, but are not limited to, the expected resource consumption of the target advertisement, the expected revenue target, and the targeting conditions of the target advertisement (i.e., the merchant's advertising requirements, such as the target gender group, age group, and advertisement display position). The specific settings can be made according to the needs and are not limited here.

[0071] Step 105: Based on advertising performance metrics, select the target ad placement combination from the ad placement combination and place the target ads into the target ad placement combination.

[0072] In this embodiment, the advertising delivery system can obtain the advertising delivery metrics of the target advertisement to match a suitable combination of target advertisements, thereby delivering the target advertisement.

[0073] It should be noted that by combining ad placements, the overall inventory utilization rate of ad placements can be improved. For example, by limiting the full utilization of inventory through low CTR placements, the overall utilization rate can be increased by 35%+.

[0074] In some embodiments, since advertising metrics may include the expected resource consumption of the target ad, the expected revenue metric value, and the targeting conditions of the target ad, an ad placement combination that meets the advertising metrics can be selected from the ad placement combination as the target ad placement combination based on the advertising metrics.

[0075] For example, a target combination of ad placements with available resources greater than the expected resource consumption of the target ad is selected based on the expected resource consumption of the target ad.

[0076] For example, based on the expected revenue metric value of the target advertisement, select ad placement combinations whose expected combined revenue metric value is greater than that expected revenue metric value.

[0077] For example, the targeting criteria for selecting an ad placement combination based on the targeting criteria of the target ad are ad placement combinations that include the targeting criteria of the target ad.

[0078] In some embodiments, delivering a target ad to a target ad placement combination includes: determining the ad traffic allocation ratio for the target ad in each ad placement in the target ad placement combination; allocating corresponding ad traffic to each ad placement in the target ad placement combination based on the ad traffic allocation ratio; and delivering the target ad to each ad placement in the target ad placement combination based on the ad traffic of each ad placement in the target ad placement combination.

[0079] The ad traffic allocation ratio is used to indicate the proportion of ad traffic allocated to the target ad. This ad traffic allocation ratio can be a preset fixed ratio or a dynamically changing allocation ratio obtained according to preset constraints. The specific setting can be determined according to the needs and is not limited here.

[0080] Specifically, the advertising delivery system can include multiple ad slots into a unified traffic allocation pool and dynamically adjust the traffic weight of each ad slot in the pool according to preset constraints, i.e., the ad traffic allocation ratio. Alternatively, it can use a preset fixed ratio to allocate traffic to each ad slot in the target ad slot combination, such as a fixed ratio of 1:1 or 2:1, so as to allocate corresponding ad traffic to different ad slots based on the ad traffic allocation ratio of each ad slot, so as to deliver the target ads.

[0081] Among them, the preset constraints can be based on the matching degree with the advertising placement indicators to assign corresponding weights. For example, within the target ad placement combination, the ad placement with a higher matching degree with the targeting conditions of the advertising placement indicators will have a larger proportion of ad traffic allocated, while the ad placement with a lower matching degree with the targeting conditions of the advertising placement indicators will have a smaller proportion of ad traffic allocated.

[0082] Alternatively, the preset constraints can be adaptively adjusted based on the advertising progress after the target ad is launched. Since progress imbalances are prone to occur in scenarios with fluctuating traffic, the proportion of ad traffic delivered to the corresponding ad slot when a certain progress is lagging behind can be adaptively adjusted based on the advertising progress.

[0083] Specifically, after delivering the target ad to each ad in the target ad group based on the ad traffic of each ad in the target ad group, the process also includes: determining the ad delivery progress of the target ad after delivery; adjusting the ad traffic allocation ratio of each ad in the target ad group for the target ad according to the ad delivery progress, to obtain the adjusted ad traffic allocation ratio; and allocating corresponding ad traffic to each ad in the target ad group based on the adjusted ad traffic allocation ratio.

[0084] The advertising delivery progress is used to indicate the progress after the target advertisement is launched. This advertising delivery progress may include, but is not limited to, resource usage progress and / or revenue progress.

[0085] Optionally, the ad traffic allocation ratio of each ad slot in the target ad slot combination to the target ad can be adjusted according to the ad campaign progress. This can include: adjusting the ad traffic allocation ratio of each ad slot in the target ad slot combination to the target ad in response to a traffic adjustment event triggered by the ad campaign progress. This ad traffic allocation ratio can be adjusted according to the ad campaign progress or according to a preset adjustment method. The specific settings can be made according to the needs and are not limited here.

[0086] Among them, the traffic adjustment event is an event used to trigger an adjustment of the advertising traffic allocation ratio of each ad slot in the target ad slot combination. This traffic adjustment event can be triggered when the deviation between the current advertising delivery progress and the expected progress is identified based on the advertising delivery progress.

[0087] It is understandable that the timing of traffic adjustment events can be either real-time monitoring of ad delivery progress to trigger when conditions are met, or periodic monitoring of ad delivery progress to trigger when conditions are met periodically.

[0088] In some embodiments, the advertising delivery progress includes resource usage progress and / or revenue progress; determining the advertising delivery progress of a target ad after delivery includes: determining the resource usage progress of the target ad after delivery based on the first resource usage consumed by the target ad after delivery and the second resource usage expected to be consumed by the target ad; and / or determining the revenue progress of the target ad after delivery based on the first revenue indicator value achieved by the target ad after delivery and the second revenue indicator value expected to be achieved by the target ad.

[0089] The first resource usage refers to the resources already consumed by the target ad in at least one ad slot within the ad slot combination after the ad is placed, such as the number of impressions of the target ad.

[0090] The second resource usage refers to the resources that the target ad is expected to consume after it is launched. For example, the target impressions of the target ad can be a target defined by the merchant to which the target ad belongs.

[0091] Specifically, the resource usage progress of a target ad after it has been launched can be the ratio or difference between the first resource usage and the second resource usage, such as Impression Progress = Number of Impressions / Target Impressions.

[0092] The first revenue indicator value refers to the revenue that the target ad has achieved in at least one ad slot within the ad slot combination after the ad is placed. For example, the actual effect of the target ad (such as actual clicks, actual conversions, actual interactions, etc.).

[0093] The second revenue indicator refers to the expected revenue that the target ad will achieve after it is launched. For example, the target effect volume of the target ad. This target effect volume can be a target defined by the merchant to which the target ad belongs, or it can be the target effect volume = target exposure volume × expected revenue indicator value.

[0094] Specifically, the resource usage progress of a target ad after it has been launched can be the ratio or difference between the first resource usage and the second resource usage. For example, the impression progress (denoted as Impression Progress) is equal to the number of impressions already made / the target number of impressions.

[0095] Understandably, by aligning progress based on both resource usage and revenue dimensions and conducting synchronous monitoring, we can ensure that resource usage progress proceeds as planned while revenue progress meets targets simultaneously, achieving a balance between quantity and quality. This allows for targeted monitoring of efficiency differences among sub-ad slots within the ad mix, thereby reducing click-through rate fluctuations by 75%.

[0096] In some embodiments, in a scenario where the progress of ad delivery is monitored periodically to trigger a traffic adjustment event when conditions are met periodically, the target ad's delivery period after delivery can be divided into a preset number of time periods (e.g., the whole day can be divided into several time periods) to allocate a second resource usage and a second revenue indicator value to each time period.

[0097] Among them, the second resource usage and second revenue index value allocated to each time period can be dynamically planned based on traffic prediction, that is, based on the peak and off-peak periods when different ad positions present ads in different time periods. For example, more second resource usage can be allocated during peak periods, while the second resource usage can be appropriately reduced during off-peak periods.

[0098] For example, the second resource usage for time period t: TargetImp_{t} = TotalImp × (Forecast_{t} / ΣForecast).

[0099] Where TargetImp_{t} is the second resource usage in time period t, TotalImp is the total resource usage in the campaign period, Forecast_{t} is the predicted available resource amount in time period t (e.g., 20:00-21:00), and ΣForecast is the predicted available resource amount in the entire campaign period.

[0100] In this embodiment, the advertising delivery system can predict the amount of available resources in each time period based on a time-series prediction model. Then, it can ensure that the resource usage progress and revenue progress of the ad placement combination are in sync through a two-dimensional progress alignment mechanism.

[0101] Specifically, a time series forecasting model (such as the XGBoost model) can be used to predict the available resources for each time period within the campaign period (such as the next 24 hours) at a preset time granularity (example: 15 minutes). Based on the prediction results, the second resource usage required for the corresponding time period can be obtained. Time-sharing progress planning can be carried out based on the first and second resource usage consumed by the target advertisement to establish a two-dimensional monitoring system and maintain the progress of the two dimensions in sync.

[0102] In some embodiments, the deviation between the current advertising progress and the expected progress may include: Resource utilization deviation rate: δ_imp = (ActualImp - TargetImp) / TargetImp.

[0103] Where δ_imp is the resource usage deviation rate (such as exposure deviation rate), ActualImp is the first resource usage, and TargetImp is the second resource usage.

[0104] Profit deviation rate: δ_clk = (ActualClk - TargetClk) / TargetClk.

[0105] Where δ_clk is the revenue deviation rate (such as click deviation rate), ActualClk is the first revenue metric value, and TargetClk is the second revenue metric value.

[0106] Accordingly, traffic adjustment events can be triggered when the resource usage deviation rate and / or revenue deviation rate exceed a preset deviation threshold (e.g., ±10%).

[0107] In some embodiments, where the target ad placement combination includes at least a first category and a second category of ad placements, the target available resource amount of the first category of ad placements is greater than a first preset resource threshold and the expected revenue indicator value is less than a first preset revenue threshold, and the target available resource amount of the second category of ad placements is less than a second preset resource threshold and the expected revenue indicator value is greater than a second preset revenue threshold, the specific adjustment method for the ad traffic allocation ratio of the target ad placement combination may be to adjust the first category and the second category of ad placements in a specific direction.

[0108] Specifically, based on the progress of ad campaigns, the ad traffic allocation ratio for each ad slot in the target ad slot combination is adjusted to obtain the adjusted ad traffic allocation ratio, including: If the resource usage progress meets the first preset lag condition and the revenue progress meets the first preset improvement condition, then increase the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio; and / or If the resource utilization progress meets the second preset progress improvement condition and the revenue progress meets the second preset progress lag condition, then increase the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio; and / or If the resource usage progress meets the first preset progress lag condition and the revenue progress meets the second preset progress lag condition, then increase the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio.

[0109] In this embodiment, the direction of the advertising traffic allocation ratio of a specific category of advertising slots in the target advertising slot combination can be adjusted using a preset decision matrix based on the relative situation of resource usage progress and revenue progress.

[0110] Among them, the resource usage progress meeting the first preset progress lag condition can be that the resource usage progress is lower than the expected resource usage progress, such as a negative resource usage deviation rate.

[0111] Among them, the revenue progress meeting the first preset progress improvement condition can be that the revenue progress is higher than the expected revenue progress, such as a positive revenue deviation rate.

[0112] Among them, the resource usage progress meeting the second preset progress improvement condition can be that the resource usage progress is higher than the expected resource usage progress, such as a positive resource usage deviation rate.

[0113] Among them, the revenue progress meeting the second preset progress lag condition can be that the revenue progress is lower than the expected revenue progress, such as a negative revenue deviation rate.

[0114] Furthermore, if the resource utilization progress meets the second preset progress improvement condition and the revenue progress meets the first preset progress improvement condition, it indicates that both exposure and click-through rate have been achieved ahead of schedule. In this case, the advertising system can optimize the target ad placement combination in the later stages, such as reducing the overall bidding competitiveness of the first category ad placements and the second category ad placements. Alternatively, if the resource utilization progress meets the first preset progress lag condition and the revenue progress meets the second preset progress lag condition, the advertising system can optimize the target ad placement combination in the later stages, such as increasing the ad traffic allocation ratio of the second category ad placements in the target ad placement combination and decreasing the ad traffic allocation ratio of the first category ad placements in the target ad placement combination.

[0115] For example, the decision matrix can be a four-quadrant decision matrix. By utilizing the deviation direction decision mechanism in the four-quadrant decision matrix, such as the adjustment method of the advertising traffic allocation ratio mentioned in this embodiment, the advertising traffic allocation ratio can be adjusted.

[0116] Understandably, by adopting a four-quadrant deviation direction decision-making mechanism, the weight adjustment direction is determined based on four different combinations of resource usage progress and revenue progress. This improves overall resource usage while ensuring stable click efficiency, achieving precise restoration of progress in both dimensions under different deviation scenarios. Furthermore, by adjusting the weight of each ad slot within the ad slot combination, the distribution of ad traffic in subsequent time periods is influenced, thereby meeting user expectations as quickly as possible.

[0117] By using the positive and negative directions of exposure deviation rate (δ_imp) and click deviation rate (δ_clk) as two dimensions, four deviation combination states (four quadrants) are constructed. For each state, the adjustment direction of the ad traffic allocation ratio of the corresponding category of ad slots is defined. That is, when exposure lags and click lags, bid competitiveness is increased; when exposure lags and click leads, the weight of high efficiency is reduced and the weight of low efficiency is increased; when exposure leads and click lags, the weight of high efficiency is increased and the weight of low efficiency is reduced; when exposure leads and click leads, competitiveness is reduced and click efficiency stability is improved, so that the click efficiency fluctuation is reduced to ±5% and the exposure progress deviation recovery time is accelerated to one monitoring cycle (5 minutes, 15 minutes, etc.).

[0118] In some embodiments, adjusting the ad traffic allocation ratio of each ad slot in the target ad slot combination for the target ad according to the ad delivery progress to obtain the adjusted ad traffic allocation ratio includes: determining the traffic adjustment direction of each ad slot in the target ad slot combination for the target ad according to the ad delivery progress; and adjusting the ad traffic allocation ratio of each ad slot in the target ad slot combination for the target ad according to the traffic adjustment direction of each ad slot in the target ad slot combination and the preset adjustment step size parameter to obtain the adjusted ad traffic allocation ratio.

[0119] Specifically, the formula for adjusting the ad traffic allocation ratio for each ad slot in the target ad mix is ​​as follows: w' i = w i × (1 + η × Δ i ).

[0120] Among them, w' i This is the adjusted ad traffic allocation ratio, w i This is the original ad traffic allocation ratio, Δ i The direction of traffic adjustment can be determined based on increasing or decreasing the proportion of ad traffic allocated to ad slots in the above embodiments, where η is a preset adjustment step size parameter, and w i ∈ [w_min, w_max], by setting w i Introduce weight boundary constraints to prevent excessive weight skew.

[0121] Specifically, a continuous adjustment function (such as a PID controller, linear programming, etc.) can be used to adjust the proportion of ad traffic allocated to the target ad in each ad slot in the target ad slot combination. The direction of this traffic adjustment can still be determined based on the increase or decrease of the ad traffic allocation ratio of the ad slot in the above embodiment.

[0122] For example, such as Figure 2 As shown, the advertising delivery system can be equipped with an inventory management and progress alignment module (denoted as Progress Alignment). Based on this module, the system can monitor the progress of the target ad after it has been launched and adjust the ad traffic allocation ratio of each ad slot in the target ad combination based on the progress.

[0123] In some embodiments, after delivering a target ad to a target ad placement combination, the method further includes: if an ad request for a target ad placement within the target ad placement combination is received, determining the current ad traffic of the target ad placement based on the ad traffic allocation ratio corresponding to the target ad placement combination; and controlling the delivery of the target ad to the target ad placement based on the current ad traffic of the target ad placement.

[0124] In this embodiment, if an ad request is received for a target ad slot within a target ad slot combination, the current ad traffic of the target ad slot is determined based on the ad traffic allocation corresponding to the target ad slot combination. If the current ad traffic of the target ad slot meets the delivery conditions, the request is routed to the target ad slot and the delivery is executed.

[0125] For example, the process between receiving an ad request and ad delivery may include the following steps: Step 1, T=0ms: The ad request arrives. The request parsing module extracts the user ID=U123, device=Mobile, page type=news, and time=14:30.

[0126] Step 2, T=1ms: Based on the ad request, query the ad slot index (i.e., the ad identifier mentioned above), match the order placed by a certain advertiser, and the corresponding merged slot F001 (formed by merging ad slot SA (H type, CTR=0.06, inventory=8000) and SB (L type, CTR=0.015, inventory=60000)).

[0127] Step 3, T=3ms: Read the ad traffic allocation ratio corresponding to the current ad slot SA as w_A=0.65, and the ad traffic allocation ratio corresponding to the ad slot SB as w_B=0.35 (updated by the progress alignment module 5 minutes ago, currently in a situation of exposure ahead and click lagging).

[0128] Step 4, T=4ms: If the current ad request is SA, return the ad creative and complete the impression record; if it is SB ad placement, and the ad traffic allocation ratio of SB ad placement is small or 0, then do not return the ad creative. Step 5, after T=5 minutes: The progress alignment module calculates δ_imp=+8% (slightly ahead, not triggered) and δ_clk=-3% (slightly behind, not triggered), and maintains the current weight.

[0129] For example, such as Figure 2 As shown, the advertising delivery system can be equipped with an allocation execution module (denoted as Real-timeAllocation). The advertising delivery system can receive advertising requests sent to it by the platform based on this allocation execution module, that is, requests to display advertisements in specific ad slots. After receiving the advertising request, the advertising delivery system can parse the advertising request, such as parsing signals such as user ID, device information, page context, and request time.

[0130] Then, the ad placement combination associated with the ad request is analyzed, that is, the ad placement combination that matches the user targeting conditions associated with the ad request and the specific ad placements within the ad placement combination, and traffic allocation is performed, that is, based on the ad traffic ratio of the specific ad placements within the ad placement combination and the HWM algorithm of the ads associated with the ad request. For example, if ad placement A has a weight of 30% and ad placement B has a weight of 70%, then the exposure progress of A and B will be controlled at 3:7 by the HWM algorithm.

[0131] Finally, the request is routed to the selected ad slot to execute the ad delivery.

[0132] Optionally, upon receiving an ad request, the optimal sub-ad slot can be selected for placement based on current efficiency estimates.

[0133] In some embodiments, an end-to-end online allocation system architecture can be designed for the advertising delivery system. This architecture integrates a traffic prediction module, a fusion decision module, an inventory management module, an allocation execution module, and a progress alignment module for online advertising delivery. Each module achieves efficient decoupling and collaboration through an asynchronous message middleware (such as Kafka), supporting a real-time traffic allocation closed loop with millisecond-level response.

[0134] Understandably, compared to operating each module independently, this embodiment designs an end-to-end integrated module for converged scenarios, achieves millisecond-level real-time closed loop through asynchronous message middleware, minimizes system response latency (P99 < 50ms), and enables real-time operation. This allows for the integration of converged decision-making capabilities within the supplier platform, eliminates cross-system communication delays, supports real-time online decision-making, and ensures decoupling of functional units through asynchronous message middleware.

[0135] Furthermore, all modules within this advertising delivery system operate collaboratively within a millisecond-level response cycle for a single advertising request, forming an end-to-end online real-time closed loop. In addition, a hybrid mode combining offline pre-computation and online real-time execution is employed (e.g., offline fusion decision-making and online allocation execution).

[0136] The data interface definitions for each module are as follows: Traffic prediction module output (ad slot ID, predicted inventory vector, efficiency metric); Fusion decision module output (fusion slot ID, sub-ad slot mapping, fusion parameters); Inventory management module output (time-sharing target schedule); Schedule alignment module output (weight adjustment vector); Allocation execution module input (user request signal) output (sub-ad slot ID, delivery result).

[0137] It should be noted that during the 30-day test period, covering 12 ad placements (including 4 H-type, 5 L-type, and 3 B-type, covering various ad placement types such as homepage banners, rewarded videos, and splash screens), with an average daily traffic of approximately 5 million page views, and using an A / B controlled experiment design, with the experimental group using integrated allocation and the control group using independent ad placement management, the above ad placement method can improve the overall inventory utilization rate by more than 35%, control the click efficiency fluctuation within ±5%, and achieve an end-to-end response latency (P99) of less than 50 milliseconds.

[0138] In some embodiments, such as Figure 3 As shown, after an ad request arrives, the following steps can be taken: Step 1: Traffic prediction, which involves predicting the exposure and CTR of each ad placement to obtain the traffic prediction results; Step 2: Based on the traffic forecast results, classify the ad placements. That is, classify the ad placements into H-type, L-type and B-type based on efficiency indicators and inventory capacity. Then determine whether there is a combination that can be integrated between H-type and L-type. If not, the ad placements will be deployed independently. If so, proceed to step 3. Step 3: Perform fusion condition verification, that is, determine the ad placements that can be matched under the joint judgment of the three constraints: CTR constraint (i.e., efficiency constraint), inventory constraint, and audience similarity (i.e., audience overlap constraint). Step 4: Based on the available ad slots, create virtual merged ad slots (i.e., ad slot combinations), assign a unified identifier to each virtual merged ad slot, and establish a mapping relationship between virtual merged ad slots and different identifiers; Step 5: Based on the dual-dimensional objectives of "exposure progress" (i.e., resource usage progress) and "click speed" (i.e. revenue progress), conduct time-sharing progress planning. For example, divide the time into 96 time periods with a 15-minute time interval, and dynamically allocate resources to the 96 time periods. Step 6: Real-time allocation and execution based on ad requests. That is, the ad is routed to the sub-ad slot using a weighted random algorithm. Then, based on the two-dimensional target, it is determined whether the two-dimensional deviation exceeds ±10%. If so, proceed to step 7. If not, continue ad delivery until the ad delivery is completed, and then update the statistical results. Step 7: Adjust the ad traffic allocation ratio of the ad slots using a four-quadrant matrix, and then return to step 6, which uses a weighted random algorithm to route the ads to specific sub-ad slots for ad delivery.

[0139] As can be seen from the above, by obtaining the target available resources of multiple ad slots and the expected revenue indicators of each ad slot; based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, which includes at least two ad slots with different resource and revenue attributes; the ad placement indicators of the target ads to be placed are obtained; based on the ad placement indicators, a target ad slot combination is selected from the ad slot combinations, and the target ads are placed on the target ad slot combination. Thus, by using ad slot combinations, the server can perform unified indexing and scheduling of multi-source heterogeneous data when responding to high-concurrency ad requests, avoiding a large amount of repetitive calculations and redundant signaling interactions, thereby greatly reducing data processing pressure and system resource consumption. Furthermore, it can enable the server to synchronize and align the placement status of multiple dimensions in real time at the global level, thereby improving the overall resource utilization of the system.

[0140] This embodiment also provides an advertising delivery device, which can be integrated into a terminal device. For example, such as... Figure 4 As shown, the advertising delivery device may include: The information acquisition module 401 is used to acquire the target available resources of multiple ad slots and the expected revenue index value of each ad slot. The ad placement combination module 402 is used to combine and associate at least two ad placements with complementary resource and revenue attributes based on the target available resource amount and expected revenue index value of each ad placement, so as to obtain at least one ad placement combination. The ad placement combination includes at least two ad placements with different resource and revenue attributes. The metrics acquisition module 404 is used to acquire the advertising metrics of the target advertisement to be delivered; The ad delivery module 405 is used to select a target ad placement combination from the ad placement combination based on ad delivery metrics, and to deliver the target ad to the target ad placement combination.

[0141] In some embodiments, serving targeted ads to a combination of targeted ad placements includes: Determine the proportion of ad traffic allocated to each ad slot in the target ad mix for the target ad; Based on the ad traffic allocation ratio, allocate corresponding ad traffic to each ad slot in the target ad slot combination; Based on the ad traffic of each ad slot in the target ad slot combination, target ads are delivered to each ad slot in the target ad slot combination.

[0142] In some embodiments, after delivering target ads to each ad slot in the target ad slot combination based on the ad traffic of each ad slot in the target ad slot combination, the method further includes: Determine the ad delivery schedule after the target ad is launched; Based on the advertising campaign progress, adjust the ad traffic allocation ratio for each ad slot in the target ad slot combination to obtain the adjusted ad traffic allocation ratio. Based on the adjusted ad traffic allocation ratio, return to the step of allocating corresponding ad traffic to each ad slot in the target ad slot combination based on the ad traffic allocation ratio.

[0143] In some embodiments, the advertising delivery progress includes resource usage progress and / or revenue progress; Determine the ad delivery schedule after the target ad is launched, including: Based on the first resource usage consumed by the target ad after launch, and the expected second resource usage, determine the resource usage progress of the target ad after launch; and / or Based on the first revenue indicator value achieved by the target ad after its launch, and the second revenue indicator value expected to be achieved by the target ad, the revenue progress of the target ad after its launch is determined.

[0144] In some embodiments, the target ad placement combination includes at least ad placements of a first category and a second category, wherein the target available resource amount of the ad placements of the first category is greater than a first preset resource threshold and the expected revenue index value is less than a first preset revenue threshold, and the target available resource amount of the ad placements of the second category is less than a second preset resource threshold and the expected revenue index value is greater than a second preset revenue threshold.

[0145] In some embodiments, the ad traffic allocation ratio for each ad slot in the target ad slot combination is adjusted according to the ad delivery progress to obtain the adjusted ad traffic allocation ratio, including: If the resource usage progress meets the first preset lag condition and the revenue progress meets the first preset improvement condition, then increase the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio; and / or If the resource utilization progress meets the second preset progress improvement condition and the revenue progress meets the second preset progress lag condition, then increase the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio; and / or If the resource usage progress meets the first preset progress lag condition and the revenue progress meets the second preset progress lag condition, then increase the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination and decrease the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination to obtain the adjusted ad traffic allocation ratio.

[0146] In some embodiments, the ad traffic allocation ratio for each ad slot in the target ad slot combination is adjusted according to the ad delivery progress to obtain the adjusted ad traffic allocation ratio, including: Based on the progress of the advertising campaign, determine the direction of traffic adjustment for each ad placement in the target ad combination for the target ad. Based on the traffic adjustment direction of each ad slot in the target ad slot combination and the preset adjustment step size parameter, the ad traffic allocation ratio of each ad slot in the target ad slot combination for the target ad is adjusted to obtain the adjusted ad traffic allocation ratio.

[0147] In some embodiments, after serving targeted ads to a combination of targeted ad placements, the method further includes: If an ad request is received for a target ad slot within a target ad slot combination, the current ad traffic for the target ad slot is determined based on the ad traffic allocation corresponding to the target ad slot combination. Based on the current ad traffic of the target ad slot, control the delivery of target ads to the target ad slot.

[0148] In some embodiments, based on the target available resources and expected revenue metrics of each ad placement, at least two ad placements with complementary resource and revenue attributes are combined and associated to obtain at least one ad placement combination, including: Based on the target available resources and expected revenue metrics for each ad slot, determine the ad category for each ad slot; Under the premise of satisfying the preset ad placement combination constraints, at least two ad placements with complementary resources and revenue attributes are combined and associated based on the ad category of each ad placement to obtain at least one ad placement combination. The preset ad placement combination constraints may include at least one of the following: The expected combined revenue index value corresponding to the ad placement combination is greater than or equal to the preset combined revenue index threshold. The available resource amount of the target combination corresponding to the ad placement combination is greater than or equal to the preset threshold of available resource amount for the combination. The audience overlap of each ad placement in the ad placement combination is less than the preset overlap threshold.

[0149] As can be seen from the above, by obtaining the target available resources of multiple ad slots and the expected revenue indicators of each ad slot; based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, which includes at least two ad slots with different resource and revenue attributes; the ad placement indicators of the target ads to be placed are obtained; based on the ad placement indicators, a target ad slot combination is selected from the ad slot combinations, and the target ads are placed on the target ad slot combination. Thus, by using ad slot combinations, the server can perform unified indexing and scheduling of multi-source heterogeneous data when responding to high-concurrency ad requests, avoiding a large amount of repetitive calculations and redundant signaling interactions, thereby greatly reducing data processing pressure and system resource consumption. Furthermore, it can enable the server to synchronize and align the placement status of multiple dimensions in real time at the global level, thereby improving the overall resource utilization of the system.

[0150] Accordingly, this application also provides an electronic device, which can be a terminal, such as a smartphone, tablet computer, laptop computer, touch screen, game console, personal computer (PC), personal digital assistant (PDA), or other terminal device. Alternatively, the electronic device can be a server.

[0151] like Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 500 includes a processor 501 with one or more processing cores, a memory 502 with one or more computer-readable storage media, and a computer program stored on the memory 502 and executable on the processor. The processor 501 and the memory 502 are electrically connected. Those skilled in the art will understand that the electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0152] The processor 501 is the control center of the electronic device 500. It connects various parts of the electronic device 500 via various interfaces and lines. By running or loading software programs and / or units stored in the memory 502, and by calling data stored in the memory 502, it executes various functions and processes data of the electronic device 500, thereby providing overall monitoring of the electronic device 500. The processor 501 can be a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), etc., and can implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of this application.

[0153] In this embodiment, the processor 501 in the electronic device 500 loads the instructions corresponding to the processes of one or more applications into the memory 502 according to the following steps, and the processor 501 runs the applications stored in the memory 502 to realize various functions, such as: Obtain the target available resources for multiple ad placements, and the expected revenue metrics for each ad placement; Based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination. The ad slot combination includes at least two ad slots with different resource and revenue attributes. Obtain the advertising metrics for the target ads to be delivered; Based on advertising performance metrics, select target ad placement combinations from ad placement combinations and deliver target ads to these target ad placement combinations.

[0154] Therefore, the electronic device 500 provided in this embodiment can bring the following technical effects: reduce the data processing pressure and system resource overhead of the server when responding to high-concurrency advertising requests, and improve the overall resource utilization of the system.

[0155] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0156] Optional, such as Figure 5 As shown, the electronic device 500 also includes: a touch display screen 503, a radio frequency circuit 504, an audio circuit 505, an input unit 506, and a power supply 507. The processor 501 is electrically connected to the touch display screen 503, the radio frequency circuit 504, the audio circuit 505, the input unit 506, and the power supply 507. Those skilled in the art will understand that... Figure 5 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0157] The touch display screen 503 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The touch display screen 503 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Optionally, the display panel can be configured using a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar technologies. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program according to the operation commands. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 501. It can also receive and execute commands from the processor 501. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 501 to determine the type of touch event. Subsequently, the processor 501 provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the touch display screen 503 to achieve input and output functions. However, in some embodiments, the touch panel and the touch display screen 503 can be implemented as two independent components to achieve input and output functions. That is, the touch display screen 503 can also be used as part of the input unit 506 to achieve input functions.

[0158] The radio frequency circuit 504 can be used to transmit and receive radio frequency signals to establish wireless communication with network devices or other electronic devices, and to transmit and receive signals with network devices or other electronic devices.

[0159] Audio circuitry 505 can be used to provide an audio interface between a user and an electronic device via a speaker and a microphone. Audio circuitry 505 converts received audio data into electrical signals, transmits them to the speaker, and the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuitry 505, converted back into audio data, and then processed by processor 501 before being transmitted via radio frequency circuitry 504 to, for example, another electronic device, or output to memory 502 for further processing. Audio circuitry 505 may also include an earphone jack to facilitate communication between peripheral headphones and electronic devices.

[0160] The input unit 506 can be used to receive input numbers, characters, or user characteristic information (such as fingerprints, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control.

[0161] Power supply 507 is used to supply power to various components of electronic device 500. Optionally, power supply 507 can be logically connected to processor 501 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 507 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0162] although Figure 5 As not shown in the diagram, the electronic device 500 may also include a camera, sensor, wireless fidelity module, Bluetooth module, etc., which will not be described in detail here.

[0163] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0164] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0165] Therefore, embodiments of this application provide a computer-readable storage medium storing multiple computer programs that can be loaded by a processor to execute any of the advertising delivery methods provided in this application. The computer program can execute the steps of the following advertising delivery method: Obtain the target available resources for multiple ad placements, and the expected revenue metrics for each ad placement; Based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination. The ad slot combination includes at least two ad slots with different resource and revenue attributes. Obtain the advertising metrics for the target ads to be delivered; Based on advertising performance metrics, select target ad placement combinations from ad placement combinations and deliver target ads to these target ad placement combinations.

[0166] As can be seen, the computer program can be loaded by the processor to execute any of the advertising delivery methods provided in the embodiments of this application, thereby bringing the following technical effects: reducing the data processing pressure and system resource overhead of the server when responding to high-concurrency advertising requests, and improving the overall resource utilization of the system.

[0167] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0168] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0169] Since the computer program stored in the computer-readable storage medium can execute any of the advertising delivery methods provided in the embodiments of this application, it can achieve the beneficial effects that any of the advertising delivery methods provided in the embodiments of this application can achieve, as detailed in the preceding embodiments, and will not be repeated here.

[0170] According to one aspect of this application, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the methods provided in the various optional implementations of the above embodiments.

[0171] In the above embodiments of the advertising delivery device, computer-readable storage medium, electronic device, and computer program product, the descriptions of each embodiment have different focuses. Parts not described in detail in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes and beneficial effects of the advertising delivery device, computer-readable storage medium, computer program product, electronic device, and their corresponding units described above can be referred to the description of the advertising delivery method in the above embodiments, and will not be repeated here.

[0172] The above provides a detailed description of an advertising delivery method, apparatus, electronic device, computer-readable storage medium, and computer program product provided in the embodiments of this application. 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 core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An advertisement distribution method characterized by comprising: The method includes: Obtain the target available resources for multiple ad placements, and the expected revenue metrics for each ad placement; Based on the target available resources and expected revenue index value of each of the aforementioned ad slots, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, wherein the ad slot combination includes at least two ad slots with different resource and revenue attributes. Obtain the advertising metrics for the target ads to be delivered; Based on the advertising delivery metrics, a target ad placement combination is selected from the ad placement combinations, and the target ad is delivered to the target ad placement combination.

2. The advertisement distribution method according to claim 1, wherein The delivery of the target advertisement to the target ad placement combination includes: Determine the proportion of ad traffic allocated to each ad slot in the target ad slot combination for the target ad; Based on the advertising traffic allocation ratio, the corresponding advertising traffic is allocated to each advertising position in the target advertising position combination; Based on the advertising traffic of each ad slot in the target ad slot combination, the target ad is delivered to each ad slot in the target ad slot combination.

3. The advertising placement method as described in claim 2, characterized in that, After delivering the target advertisement to each ad slot in the target ad slot combination based on the ad traffic of each ad slot in the target ad slot combination, the method further includes: Determine the ad delivery schedule after the target ad is launched; Based on the advertising delivery progress, adjust the advertising traffic allocation ratio of each ad slot in the target ad slot combination for the target ad to obtain the adjusted advertising traffic allocation ratio; Based on the adjusted advertising traffic allocation ratio, corresponding advertising traffic is allocated to each advertising position in the target advertising position combination.

4. The advertising placement method as described in claim 3, characterized in that, The advertising delivery progress includes resource utilization progress and / or revenue progress; Determining the ad delivery progress of the target ad after it has been launched includes: Based on the first resource usage consumed by the target ad after its launch, and the second resource usage expected to be consumed by the target ad, determine the resource usage progress of the target ad after its launch; and / or Based on the first revenue indicator value achieved by the target advertisement after its launch, and the second revenue indicator value expected to be achieved by the target advertisement, the revenue progress of the target advertisement after its launch is determined.

5. The advertising placement method as described in claim 4, characterized in that, The target ad placement combination includes at least a first category and a second category of ad placements. The target available resource amount of the ad placements in the first category is greater than a first preset resource threshold, and the expected revenue index value is less than a first preset revenue threshold. The target available resource amount of the ad placements in the second category is less than a second preset resource threshold, and the expected revenue index value is greater than a second preset revenue threshold.

6. The advertising placement method as described in claim 5, characterized in that, The step of adjusting the ad traffic allocation ratio for each ad slot in the target ad slot combination based on the ad delivery progress, to obtain the adjusted ad traffic allocation ratio, includes: If the resource usage progress meets the first preset progress lag condition and the revenue progress meets the first preset progress improvement condition, then the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination is increased, and the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination is decreased, to obtain the adjusted ad traffic allocation ratio; and / or If the resource usage progress meets the second preset progress improvement condition and the revenue progress meets the second preset progress lag condition, then the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination is increased, and the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination is decreased, to obtain the adjusted ad traffic allocation ratio; and / or If the resource usage progress meets the first preset progress lag condition and the revenue progress meets the second preset progress lag condition, then the ad traffic allocation ratio of the second category of ad slots in the target ad slot combination is increased, and the ad traffic allocation ratio of the first category of ad slots in the target ad slot combination is decreased, to obtain the adjusted ad traffic allocation ratio.

7. The advertising placement method according to any one of claims 1 to 6, characterized in that, Based on the target available resources and expected revenue indicators of each ad slot, at least two ad slots with complementary resource and revenue attributes are combined and associated to obtain at least one ad slot combination, including: Based on the target available resources and expected revenue indicators of each ad slot, the ad category of each ad slot is determined; Under the condition of satisfying the preset ad slot combination constraints, at least two ad slots with complementary resources and revenue attributes are combined and associated based on the ad category of each ad slot to obtain at least one ad slot combination. The preset ad placement combination constraints may include at least one of the following: The expected combined revenue index value corresponding to the ad placement combination is greater than or equal to the preset combined revenue index threshold. The available resource amount of the target combination corresponding to the ad placement combination is greater than or equal to the preset threshold of available resource amount of the combination. The audience overlap of each ad slot in the ad slot combination is less than a preset overlap threshold.

8. An advertising delivery device, characterized in that, The device includes: The information acquisition module is used to acquire the target available resource quantity of multiple ad slots, as well as the expected revenue index value of each ad slot; An ad placement combination module is used to combine and associate at least two ad placements with complementary resource and revenue attributes based on the target available resource amount and the expected revenue index value of each ad placement, to obtain at least one ad placement combination, wherein the ad placement combination includes at least two ad placements with different resource and revenue attributes. The metrics acquisition module is used to acquire the advertising metrics for the target ads to be delivered. The advertising delivery module is used to select a target ad placement combination from the ad placement combination based on the advertising delivery metrics, and deliver the target ad to the target ad placement combination.

9. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing multiple instructions; the processor loads instructions from the memory to perform the steps of the advertising delivery method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to perform the steps of the advertising delivery method as described in any one of claims 1 to 7.

Images